Molecular mechanisms involved in the regulation of agouti-related and Y by endocrine disrupting chemical bisphenol A in hypothalamic neurons

by

Neruja Loganathan

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Physiology University of Toronto

© Copyright by Neruja Loganathan 2021

Molecular mechanisms involved in the regulation of agouti-related peptide and by endocrine disrupting chemical bisphenol A in hypothalamic neurons

Neruja Loganathan

Doctor of Philosophy

Department of Physiology University of Toronto

2021 Abstract

Bisphenol A (BPA), a ubiquitous endocrine disrupting chemical found in plastics and receipts, is a disruptor of reproductive function and is a known ‘obesogen’ as it is linked to increased body mass index in humans and leads to weight gain in animal models. The houses orexigenic NPY/AgRP neurons, which integrate peripheral hormones and nutritional signals, to increase food intake and decrease energy expenditure. NPY neurons are also afferent regulators of the hypothalamic-pituitary gonadal axis, and thus reproductive function. This thesis investigated whether the NPY/AgRP neurons, and particularly Npy and Agrp expression, are altered by BPA. We hypothesized that BPA increases Npy and Agrp gene expression in hypothalamic neurons and that this effect is mediated through nuclear activation, induction of cellular stress and subsequent transcription factor activation or circadian dysregulation. We demonstrated that BPA increased Agrp mRNA expression in mHypoA-59 and mHypoE-41 cells. Inhibition of AMPK and knock-down of transcription factor ATF3 prevented the BPA-mediated increase in Agrp expression in the mHypoA-59 cells. ATF3 was also required for BPA-mediated increase in Npy in the mHypoE-41 cells. We also described subpopulation- specific changes in Npy expression in response to BPA. While BPA induced Npy expression in

ii mHypoA-59, -2/12 and mHypoE-41, -42 neurons, Npy expression was downregulated in mHypoE-46 and -44 neurons. Inhibition of AMPK with compound C or oxidative stress with antioxidants and vitamin B6 prevented the BPA-mediated induction in Npy in the mHypoA-59 cells, whereas antagonism of ERß or GPER prevented the decrease in Npy mRNA in the mHypoE-46 cells. Finally, we showed that circadian gene dysregulation occurred with BPA exposure and using hypothalamic cell lines lacking BMAL1, demonstrated that the BPA- mediated induction of Npy, but not Agrp, required BMAL1. Accordingly, treatment with BPA increased BMAL1 binding to the Npy promoter. These findings illustrate distinct mechanisms responsible for the BPA-mediated changes in appetite-increasing Npy and Agrp gene expression, suggesting that NPY/AgRP neurons are susceptible to the endocrine disrupting effects of BPA.

Furthermore, we describe potential targetable pathways to combat the obesogenic or reproductive dysfunction-inducing effects of BPA at the hypothalamic level.

iii

Acknowledgements

First and most importantly, I would like to thank my supervisor, Dr. Denise Belsham. You have helped me develop my skills as a scientist, taught me the importance of always asking questions and instilled in me a love for neuroendocrinology. You have been compassionate and motivating and I cannot imagine having done my PhD anywhere else. Your passion for encouraging women in science is truly remarkable, and I will forever be grateful for the opportunities you have given me and the continuous support you have provided me.

I would also like to thank my supervisory committee members - Dr. Mark Palmert, Dr. Carolyn Cummins and Dr. Amira Klip. The challenging questions you asked and the careful input you provided was essential for the development and completion of my project. Thank you for helping me become a better scientist. I also wish to thank Dr. Debby Kurrasch, Dr. Patricia Brubaker and Dr. Andy Babwah for their comments and edits on my thesis. Thank you to the Department of Physiology - from the administrative staff that have been nothing but kind and helpful to the neighbouring labs that have always lent support when I needed to borrow equipment.

The last five years would not have made such an impact on my life if it were not for all the members of the Belsham Lab. You have helped me with my experiments, you have provided ample suggestions for my project and you have made me laugh a million times in between. I am very thankful for having gotten to know each one of you. To Jenn - thank you for teaching me every technique I know and helping me troubleshoot my experiments. You always made time to help someone when they needed it, which is so admirable. To Erika - thank you for navigating the beginning stages of graduate school with me. Your enthusiasm for figuring things out and learning something new made working with you so fun. To Andy and Calvin - thank you for making everyday in the lab so joyful. To Emma - from the countless experiments, to looking up the answer to a question or editing my writing, you have been a tremendous support throughout my entire degree. Thank you for always helping me, but most importantly, thank you for your friendship.

Finally, I would like to thank my friends and family for their constant encouragement and emotional support. Thank you for believing in me. To my mother, I cannot imagine where I would be without you. Thank you for absolutely everything.

iv

Table of Contents

Acknowledgements ...... iv

Table of Contents ...... v

List of Tables and Figures...... xi

List of Abbreviations ...... xiv

List of Appendices ...... xviii

Chapter 1 Introduction...... 1

1 Introduction ...... 2

1.1 Obesity and reproductive dysfunction ...... 2

1.2 Endocrine disrupting chemicals ...... 3

1.2.1 General characteristics ...... 3

1.2.2 Bisphenol A: Sources and Concentrations ...... 5

1.2.3 BPA disrupts energy homeostasis ...... 7

1.2.3.1 Humans ...... 7

1.2.3.2 Animal and in vitro models ...... 8

1.2.4 BPA induces reproductive dysfunction ...... 10

1.3 Hypothalamic control of energy homeostasis and reproduction...... 10

1.3.1 Energy homeostasis is controlled by NPY/AgRP and POMC neurons ...... 10

1.3.1.1 NPY and AgRP are potent orexigens ...... 11

1.3.1.2 NPY and AgRP regulate metabolism independent of food intake ...... 12

1.3.2 Reproduction is controlled by GnRH neurons ...... 13

1.3.2.1 Afferent regulators of GnRH neurons ...... 14

1.3.3 Regulation of Npy and Agrp transcription and secretion ...... 15

1.3.3.1 Regulation of Npy ...... 16

1.3.3.2 Regulation of Agrp ...... 18

1.4 Hypothalamic control of circadian rhythms ...... 21 v

1.4.1 Circadian involvement in Npy regulation ...... 23

1.5 Effects of BPA in the hypothalamus ...... 23

1.6 Signaling pathways activated by BPA ...... 25

1.6.1 Hormone and nuclear receptors ...... 26

1.6.2 Non-nuclear receptor pathways including induction of cellular stress ...... 27

1.7 Hypothalamic cell models to study BPA-induced dysregulation ...... 29

1.8 Hypothesis and Aims ...... 32

Chapter 2 Materials and Methods ...... 34

2 Materials and Methods ...... 35

2.1 Cell culture and reagents ...... 35

2.1.1 Immortalized cell lines ...... 35

2.1.2 Primary culture...... 35

2.1.3 Preparation of mHypoA-Bmal1-WT and mHypoA-Bmal1-KO cell lines ...... 35

2.1.4 BPA treatment ...... 36

2.2 BPA content enzyme-linked immunosorbent assay (ELISA) ...... 38

2.3 Inhibitors, antagonists and supplements ...... 38

2.3.1 Inhibitors ...... 39

2.3.2 Steroid receptor antagonists ...... 39

2.3.3 Anti-oxidant media and supplements...... 40

2.4 Quantitative RT-PCR ...... 41

2.5 Western Blotting ...... 44

2.5.1 Cell treatments ...... 44

2.5.1.1 JNK and ERK signaling ...... 44

2.5.1.2 CHOP and ATF3 levels ...... 44

2.5.1.3 AMPK signaling ...... 44

2.5.1.4 Validation of BMAL1-KO ...... 44 vi

2.5.2 Blotting ...... 44

2.6 siRNA knockdown ...... 45

2.7 Esr1 overexpression ...... 47

2.8 In silico promoter analysis and Chromatin immunoprecipitation (ChIP) ...... 47

2.8.1 ChIP for ATF3 ...... 47

2.8.2 ChIP for BMAL1 ...... 48

2.9 Statistical Analysis ...... 49

Chapter 3 Bisphenol A induces Agrp gene expression in hypothalamic neurons through a mechanism involving ATF3 ...... 50

3 Bisphenol A induces Agrp gene expression in hypothalamic neurons through a mechanism involving ATF3 ...... 51

3.1 Abstract ...... 51

3.2 Introduction ...... 51

3.3 Results ...... 53

3.3.1 BPA increases Agrp mRNA levels in several hypothalamic cell lines ...... 53

3.3.2 BPA increases transcription of Agrp mRNA and associated transcription factors ...... 56

3.3.3 Knockdown of ATF3 blocks the BPA-induced increase in Agrp and pre-Agrp mRNA ...... 58

3.3.4 BPA does not increase ATF3 binding to Agrp regulatory elements ...... 60

3.3.5 BPA induces EndR stress, JNK, ERK and AMPK, but does not require EndR stress, JNK or ERK to increase Agrp mRNA ...... 61

3.3.6 ATF3 is also involved in the BPA-mediated increase in Npy expression ...... 64

3.4 Discussion ...... 65

Chapter 4 BPA differentially regulates Npy expression in hypothalamic neurons through a mechanism involving oxidative stress ...... 72

4 BPA differentially regulates Npy expression in hypothalamic neurons through a mechanism involving oxidative stress ...... 73

4.1 Abstract ...... 73

vii

4.2 Introduction ...... 73

4.3 Results ...... 75

4.3.1 BPA differentially alters Npy expression in subpopulations of hypothalamic neurons ...... 75

4.3.2 Steroid receptor antagonists, G15 and PHTPP, prevent BPA-induced decrease in Npy in the mHypoE-46 cells ...... 78

4.3.3 receptor mRNA levels are differentially regulated with BPA treatment ...... 80

4.3.4 AMPK inhibition prevents BPA-mediated upregulation in Npy expression ...... 83

4.3.5 BPA induces oxidative stress and neuroinflammation in Npy-expressing hypothalamic neurons ...... 84

4.3.6 Antioxidant-rich media, Neurobasal A, mitigates BPA-induced changes in Npy and oxidative stress genes ...... 87

4.3.7 NAC and vitamin B6 mitigate the BPA-mediated increase in Npy expression .....89

4.3.8 Dysregulation of estrogen receptor mRNA levels is reversed in anti-oxidant rich media...... 91

4.4 Discussion ...... 92

Chapter 5 BPA alters Bmal1, Per2 and Rev-Erba mRNA and requires Bmal1 to increase Npy expression in hypothalamic neurons ...... 99

5 BPA alters Bmal1, Per2 and Rev-Erba mRNA and requires Bmal1 to increase Npy expression in hypothalamic neurons ...... 100

5.1 Abstract ...... 100

5.2 Introduction ...... 100

5.3 Results ...... 102

5.3.1 Circadian clock gene expression is altered in response to BPA exposure ...... 102

5.3.2 Putative binding sites for BMAL1-CLOCK heterodimers exist in the regulatory regions of Npy and Agrp ...... 104

5.3.3 Characterization of mHypoA-Bmal1-KO cell lines ...... 105

5.3.4 The effect of BPA on Npy expression, but not Agrp, is dependent on Bmal1 .....108

5.3.5 BPA treatment increases BMAL1 binding to the Npy promoter ...... 110

viii

5.4 Discussion ...... 111

Chapter 6 Discussion ...... 115

6 Discussion ...... 116

6.1 Summary of Findings ...... 116

6.2 Overall Discussion ...... 118

6.3 Limitations ...... 125

6.4 Future Directions ...... 129

6.4.1 What transcription factors interact with the estrogen receptors to mediate downregulation of Npy in the mHypoE-46 cells? ...... 129

6.4.2 Do microRNAs play a role in BPA-mediated Npy regulation? ...... 130

6.4.3 Does BPA alter AgRP and NPY protein levels? ...... 130

6.4.4 Does BPA lead to hypothalamic resistance to hormones? ...... 131

6.4.5 Combination treatments with BPA alternatives and fatty acids ...... 131

6.4.6 Can antioxidant supplements protect from BPA-induced weight gain? ...... 132

6.5 Conclusions ...... 132

Appendices ...... 134

Appendix A: Steroid receptor antagonists do not block the BPA mediated increase in Agrp expression ...... 134

Appendix B: Steroid and PBA positive controls ...... 136

Appendix C: The BPA-induced increase in Agrp expression is not mediated by oxidative stress...... 138

Appendix D: The effect of BPA on Npy expression in the mHypoE-41 cells is dependent on ATF3 rather than oxidative stress ...... 140

Appendix E: The effect of BPA on Npy in primary culture ...... 141

Appendix F: Transcriptional inhibition affects the BPA-mediated changes in Npy ...... 142

Appendix G: Translational inhibition affects the BPA-mediated induction of BPA of Npy and Agrp ...... 143

Appendix H: Esr2 expression is decreased in mHypoA-BMAL1-KO cells ...... 144

ix

References ...... 145

Copyright Acknowledgements...... 177

x

List of Tables and Figures

Figure 1-1 Chemical structures of E2 and BPA...... 5

Figure 1-2 Hypothalamic control of energy homeostasis ...... 13

Figure 1-3 Afferent regulation of reproductive neurons by NPY neurons ...... 15

Figure 1-4 Regulation of NPY and AgRP expression and secretion ...... 20

Figure 1-5 Transcriptional-translational feedback loop underlying generation of circadian rhythms| ...... 22

Figure 1-6 Common signaling pathways activated by BPA ...... 28

Figure 2-1 Generation of immortalized hypothalamic cell lines ...... 37

Figure 3-1 BPA increases Agrp mRNA expression in hypothalamic cell lines...... 55

Figure 3-2 BPA increases pre-Agrp mRNA levels and associated transcription factors...... 58

Figure 3-3 ATF3 knockdown prevents BPA-mediated increase in Agrp and pre-Agrp mRNA. . 60

Figure 3-4 BPA does not increase ATF3 binding to specific Agrp regulatory regions...... 61

Figure 3-5 BPA induces EndR stress, JNK, ERK and AMPK, but does not require EndR stress, JNK or ERK to increase Agrp mRNA...... 64

Figure 3-6 ATF3 knockdown attenuates the BPA-mediated increase in Npy mRNA expression.65

Figure 4-1 BPA alters Npy gene expression in hypothalamic cell lines...... 77

Figure 4-2 Steroid receptor antagonism does not abolish effect of BPA on Npy upregulation, but antagonism of GPER or ER reverses BPA-mediated downregulation of Npy expression in mHypoE-46 cells...... 80

Figure 4-3 BPA differentially alters estrogen receptor mRNA levels in hypothalamic cell lines.82

Figure 4-4 Inhibition of AMPK blocks BPA-mediated increase in Npy in mHypoA-59 cells..... 84

xi

Figure 4-5 BPA induces oxidative stress and neuroinflammation in NPY-expressing hypothalamic cell lines...... 86

Figure 4-6 Treatment with BPA in Neurobasal A media or with ROS scavengers protects cells from Npy mRNA dysregulation...... 89

Figure 4-7 N-acetylcysteine and vitamin B6 mitigate BPA-induced Npy upregulation in mHypoA-59 cells...... 90

Figure 4-8 Neurobasal A media and vitamin B6 mitigate BPA-induced changes in Esr1 and Esr2 gene expression...... 91

Figure 4-9 Summary of Npy regulation by BPA in mHypoA-59 and mHypoE-46 cells ...... 95

Figure 5-1 BPA dysregulates circadian gene expression in hypothalamic NPY/AgRP-expressing neurons...... 104

Figure 5-2 Potential BMAL1:CLOCK binding sites in the Npy and Agrp promotors...... 105

Figure 5-3 Characterization of the mHypoA-Bmal1-WT and -KO cell models...... 108

Figure 5-4 BPA upregulates Npy and Agrp in mHypoA-Bmal1-WT/F cells, whereas the upregulation in Npy is absent in mHypoA-Bmal1-KO/F and mHypoA-Bmal1-KO/M cells. .... 109

Figure 5-5 BPA increases BMAL1 binding to the Npy promotor...... 110

Figure 6-1 Summary of Findings ...... 125

Table 1-1 Characteristics of cell lines used in current study ...... 30

Table 1-2 Regulation of (A) Npy and (B) Agrp in cell lines used in current study ...... 31

Table 2-1: qPCR Primers ...... 41

Table 2-2: siRNA Duplex sequences ...... 46

Table 4-1 Expression levels (Ct) of transcriptional regulators in the six NPY-expressing cell lines ...... 81

xii

Table 6-1 Comparison of BPA-induced effects on Npy and Agrp expression in (A) embryonic- vs. adult-derived and (B) female- vs. male-derived cell lines ...... 117

xiii

List of Abbreviations ActD: actinomycin D ADP: adenosine diphosphate AgRP: agouti-related peptide AhR: aryl hydrocarbon receptor AICAR: 5-Aminoimidazole-4-carboxamide ribonucleotide AMPK: 5’ adenosine monophosphate-activated protein kinase ANOVA: analysis of variance AR: androgen receptor ARC: arcuate nucleus ARNT: aryl receptor nuclear translocator ATF: activating transcription factor ATF3: activating transcription factor 3 ATP: AVPV: anteroventral periventricular BAT: brown adipose tissue BMAL1: brain and muscle ARNT-like 1 BMI: body mass index bp: base pairs BPA: bisphenol A BPS: bisphenol S BSX: brain-specific homeobox CAMMKß: Ca2+-calmodulin-dependent protein kinase beta CAT: catalase CC: Compound C CEBP: CCAAT/enhancer binding protein CFTR: cystic fibrosis transmembrane conductance regulator ChIP: chromatin immunoprecipitation CLOCK: circadian locomotor output cycles protein kaput CNTF: ciliary neurotrophic factor CRE: cAMP response element CREB: cAMP response element-binding protein

xiv

CRY: cryptochrome CSFBS: charcoal:dextran stripped fetal bovine serum Ct: cycle at threshold Cyc: cycloheximide DDIT3/CHOP: DNA damage inducible transcript 3 protein DMEM: Dulbecco’s Modified Eagle Medium DMSO: dimethyl sulfoxide E2: 17ß- EDC: endocrine disrupting chemical Egr1: early growth response 1 ELISA: enzyme-linked immunosorbent assay EndR stress: endoplasmic reticulum stress EPA: environmental protection agency ER: estrogen receptor ERα, Esr1: estrogen receptor alpha ERß, Esr2: estrogen receptor beta ERK1/2: extracellular signal-regulated kinase 1/2 ERRγ: estrogen related receptor gamma Esrrα: estrogen-related receptor alpha FBS: fetal bovine serum FOXO1: forkhead box protein O1 FSH: follicle stimulating hormone GABA: γ-amino butyric acid GHSR: ghrelin receptor GnRH: gonadotrophin-releasing hormone GPER, GPR30: G-protein coupled estrogen receptor Gpx1: glutathione peroxidase 1 GR, Nr3c1: glucocorticoid receptor GRP78: 78-kDa glucose-regulated protein HFD: high fat diet Hmox1: heme oxygenase 1 HPG: hypothalamic-pituitary-gonadal

xv

Hspb1, Hsp27: heat shock protein beta-1 ICV: intracerebroventricular Igf1: -like growth factor IKK: IκB kinase κB Il10: interleukin 10 Il6: interleukin 6 iNOS, Nos2: inducible nitric oxide synthase IVF: in vitro fertilization IκB: inhibitor of kappa B JNK: c-Jun N-terminal kinase KISS: KISS1: kisspeptin 1 KLF4: Krüppel-like factor 4 KO: knockout LH: luteinizing hormone LKB1: liver kinase B1

MC3/MC4: 3 and 4 MDC: metabolism disrupting chemical ME: median eminence mTOR: mamalian target of rapamycin NAC: N-acetylcysteine NBA: neurobasal A NDGA: nordihydroguariaretic acid NFκB: nuclear factor kappa B nNOS: neuronal nitric oxide synthase NO: nitric oxide NPY: neuropeptide Y Y1 – Y6: neuropeptide Y receptor 1 - 6 Nrf2: nuclear factor erythroid 2-related factor 2 PBA: sodium phenylbutyrate PBS: phosphate-buffered saline Pdk4: pyruvate dehydrogenase kinase 4

xvi

PER: periods PI3K: phosphatidyl 3-kinase POMC: pro-opiomelanocortin PPARγ: peroxisome proliferator-activated receptor γ PS: penicillin-streptomycin PS1145 dihydrochloride: PS1145 PVN: paraventricular nucleus qRT-PCR: quantitative reverse transcriptase polymerase chain reaction REV-ERB: nuclear receptor subfamily 1, group D, member 1 ROR: retinoic acid-related ROS: reactive oxygen species Rpl7: 60S ribosomal protein L7 SBP: sex steroid binding protein SCN: suprachiasmatic nucleus SHBG: steroid hormone binding globulin SOD1: superoxide dismutase 1 Sp1: specificity protein 1 STAT3: Signal transducer and activator of transcription 3 TBS-T: tris-buffered saline with tween-20 TDI: tolerable daily intake Tfap2b: transcription factor activating enhancer-binding protein 2 beta Tnfa: tumor necrosis factor alpha TAK1: transforming growth factor-beta activated kinase 1 TUDCA: tauroursodeoxycholic acid UPR: unfolded protein response WHO: World Health Organization WT: wildtype XBP1: X-box binding protein 1 XRE: xenobiotic response element ZT: zeitgeber α-MSH: α-melanocyte-stimulating hormone

xvii

List of Appendices

Appendix A: Steroid receptor antagonists do not block the BPA mediated increase in Agrp expression ...... 134

Appendix B: Steroid receptor antagonist and PBA positive controls ...... 136

Appendix C: The BPA-induced increase in Agrp expression is not mediated by oxidative stress...... 138

Appendix D: The effect of BPA on Npy expression in the mHypoE-41 cells is dependent on ATF3 rather than oxidative stress ...... 140

Appendix E: The effect of BPA on Npy in primary culture ...... 141

Appendix F: Transcriptional inhibition affects the BPA-mediated changes in Npy ...... 142

Appendix G: Translational inhibition affects the BPA-mediated induction of BPA of Npy and Agrp ...... 143

Appendix H: Esr2 expression is decreased in mHypoA-BMAL1-KO cells ...... 144

xviii

Chapter 1 Introduction

1 2

1 Introduction 1.1 Obesity and reproductive dysfunction

Obesity has become a global health pandemic, tripling worldwide since the mid-1970s. Currently (2016), obesity affects over 650 million adults worldwide, and an even more staggering 1.9 billion adults are overweight (39% of adult population). Childhood overweight/obesity is also on the rise, with 340 million children affected in 2016 (1). Obesity is defined as an excess amount of fat accumulation that may negatively impact health, and is generally characterized by a weight-height index called body mass index (BMI), where a BMI ≥ 25 kg/m2 is defined as overweight and a BMI ≥ 30 kg/m2 is defined as obesity. Critically, overweight and obesity lead to the major non-communicable causes of mortality in modern times: cardiovascular disorders, diabetes mellitus, and some cancers (2). In fact, according to the World Health Organization (WHO), approximately 2.8 million adults die each year as a result of being overweight or obese (3).

In addition to the above-mentioned complications of obesity, both overweight and obesity are linked to increased reproductive disorders and infertility. Although rates of infertility are more difficult to measure due to the complexity of factors involved, up to 10% of women worldwide (4), and approximately 16% of Canadian couples (5) and 25% of couples in developing countries experience infertility (6). Male infertility and female infertility account for approximately 30% and 40% of the Canadian cases, respectively (5), and in women, infertility is ranked the 5th highest prevalent disability in low- and middle- income countries (7). While assisted reproductive technologies are advancing, a single round of in vitro fertilization (IVF) treatment can take years to begin due to long wait times and can cost between $5000 to $20,000 CAD in Ontario (8). Obese and overweight women are 3-times more likely to experience infertility than normal weight women (9) and the odds of infertility increases with increasing BMI in men (10), underscoring the crucial link between energy balance and reproductive function. Data suggests that excess fat can affect multiple points along the hypothalamic- pituitary-gonadal (HPG) axis, which regulates the reproductive system, leading to irregular menstrual cycles, anovulation, implantation difficulties and decreased testosterone and sperm counts (9, 11).

3

Overweight and obesity result from an imbalance between the amount of energy consumed compared to that expended. Genetic and environmental factors play a role; however, the recent rapid rise in obesity has been attributed to consuming calorie-dense, high fat, high sugar diets and sedentary lifestyles, rather than genetics (2). In 2002, Baillie-Hamilton proposed a new contributing factor from the observation that the increased use of industrial chemicals worldwide has occurred alongside the rise in obesity over the last 30 years (12). Although not studied specifically for their obesity-inducing effects, many of these chemical toxins were associated with obesity. This led to the “obesogen hypothesis”: the development and maintenance of energy balance and adipogenesis can be disrupted by environmental chemicals (13). With further research, many of these chemicals were shown to cause perturbations to the body’s endocrine systems and are classified as endocrine disrupting chemicals (EDCs) (14). Independent of obesity, these EDCs are also well-known disruptors of the reproductive system (14, 15). The questions then became which EDCs are obesogens, what specific effects do they have, at what concentrations and what are the mechanisms underlying these effects.

The hypothalamus is the central regulator of energy balance, sensing peripheral hormones and metabolic cues, and integrating these cues to properly regulate food intake and energy expenditure through the action of two opposing neurons, the orexigenic neuropeptide Y(NPY) /agouti-related peptide (AgRP) neurons and the anorexigenic pro-opiomelanocortin (POMC) neurons (16). NPY neurons are also afferent regulators of neurons in the HPG axis (17). Any exogenous factors, including EDCs, that affect these neurons and their peptide products can override the intricate control of energy balance and reproduction, and ultimately lead to obesity and/or reproductive dysfunction. This thesis examines the direct effects of the common EDC and obesogen, bisphenol A (BPA), on the orexigenic NPY/AgRP neurons of the hypothalamus. Specifically, the molecular mechanisms underlying the regulation of the mRNA levels of Agrp and Npy by BPA are explored and delineated.

1.2 Endocrine disrupting chemicals

1.2.1 General characteristics

EDCs are defined as chemicals that interfere with any aspect of hormone action (13). The endocrine system regulates homeostatic processes, including metabolism, reproduction, stress, growth and development, sleep cycles and temperature, amongst other homeostatic functions. It

4 consists of glands that secrete hormones, such as estrogen, , testosterone, stress hormones and insulin, into the blood stream, which then act at distant tissues through specific receptors to mediate their effects (14). Hormones circulate at low concentrations and have specific actions on their target tissues that express the appropriate hormone receptors (18, 19). Thus, this system is highly regulated and any perturbations in the action of these hormones can be detrimental. In fact, exposure to EDCs have been shown to have several consequences, including reproductive impairment, allergies, cognitive deficits, cancer and metabolic disorders (14).

There is no single mechanism by which an EDC can act and thus, there is an extensive “list” of chemicals that are debated to be EDCs and many that have not been tested as of yet. Recently, experts in the EDC field have compiled a list of key characteristics of EDCs that aim to serve as a means to systematically organize research and data in order to resolve the debate on the specific hazard levels of EDCs, particularly as protection from disorders caused by EDCs requires government imposed limits to their use (20). From this, an EDC can act as an or antagonist to hormone receptors, can alter expression, alter signal transduction or lead to epigenetic modifications in hormone responsive cells, can alter hormone synthesis, transport or body distribution, can alter hormone metabolism or alter whether hormone- producing/responsive cells undergo proliferation, differentiation or apoptosis (20).

Examples of EDCs include bisphenols, with bisphenol A (BPA) being the primary focus of this thesis, phthalates (found in cosmetics, toys and food packaging), polychlorinated biphenyls (found in industrial solvents and insulation), pesticides, phytoestrogens (found in soy and plant products), parabens and UV filters (found in personal care products and sunscreen). Many of these chemicals disrupt hormone function because their structure, often consisting of aromatic rings, resembles that of one or more endogenous hormone(s), such as 17ß-estradiol (E2) (Figure 1-1) (14, 15).

5

A E2

B BPA

Figure 1-1 Chemical structures of E2 and BPA BPA is considered an estrogenic mimic due to its two phenol rings.

Much of the controversy as to whether an EDC is hazardous to human health comes from the fact that EDCs often have non-linear dose-response curves and as such, they can act differently at different doses, each tissue or endpoint examined can be affected differently, and the onset of effects can be delayed. These characteristics, which are particularly evident for BPA, occur as a result of the fact that EDCs do not target one receptor/pathway (20). Thus, depending on the dose (21), exposure time, endpoint examined and time at which the endpoint was examined, studies report changes to a specific phenotype or gene in both directions as well as the lack of changes, complicating the decision on whether a chemical is harmful at a certain dose (14).

1.2.2 Bisphenol A: Sources and Concentrations

BPA is a synthetic chemical found in plastics, can linings, epoxy resins, dental sealant, receipt paper, automotive parts and children’s toys. Its estrogenic potential was discovered in the 1930s,

6 yet it was widely used in the plastics industry from the 1940s onwards (22, 23). Although the use of BPA has been limited partially due to government bans (i.e. banned in use of infant bottles) and partially due to voluntary limits by industry (24), BPA is still found broadly in the environment and in humans (25). Landfill leachate (untreated – 127500 ng/L, treated – 54 – 1020 ng/L) (26, 27), waste water runoff (3 – 101 ng/L) (28, 29) and canned foods (15 – 137 ng/g) (30) represent sources in Canada. Encouragingly, it was reported that in 10 out of 16 monitoring studies of surface water downstream of waste treatment plants, the concentrations of BPA have declined in the last 10 years as a result of the imposed limits; however, levels of BPA downstream of waste treatment plants remain higher than upstream, indicating that BPA contamination in waste remains an issue (27). Furthermore, Kwan et al. reported beach sand as a major source of BPA (ranging from 22 – 215,133 µg/kg) upon studying 26 sites around the world, including beaches in the United States, Europe and Asia. Remarkably, sand contamination was significantly greater than the corresponding seawater samples (31).

BPA enters the human body primarily through the oral route, the dermal route and inhalation (25). Although the oral route is the most prominent, studies in rodents are mixed as Mielke et al. indicated that a dermal dose resulted in higher blood concentrations of BPA compared to the same oral dose (32), while Marquet et al. did not report any dermal absorption (33). Nevertheless, the dermal and inhalation route of exposure is a greater concern for those who are exposed occupationally, such as in store cashiers, wait staff and plastics manufactures, where exposure can be 10-times greater than in the general population (34). BPA is detected in the urine of 90-95% of urine samples at mean concentrations of 1.16 µg/L (7 nM) in Canadian samples (35). Recently, Gerona et al. have demonstrated that older detection methods largely underestimated the levels of BPA metabolites in human urine samples, and have reported mean concentrations to be approximately 230 nM, with as high as 3 M BPA detected in human urine (36). Serum levels of BPA have been detected up to 82 nM (18.9 ng/ml) (25), and BPA is detected in breast milk, fetal blood, placental tissue, adipose tissue, liver, the brain and even in the hypothalamus (37, 38), suggesting bioaccumulation.

BPA is primarily metabolized by the liver into a glucuronidated form and was initially reported to have a half-life of 4-5 hours after ingestion. However, Stahlhut et al. reported high urinary BPA levels after subjects had fasted for 8.5 to 24 hours (39). If oral consumption of BPA is cleared quickly, fasting time should correlate to urine BPA concentrations. Thus, BPA either has

7 a longer half-life than initially thought or non-food sources also play a large role in BPA exposure levels (39). Furthermore, reportedly ‘inactive’ glucuronidated BPA can be unconjugated by ß-glucuronidase and reabsorbed into the circulation or transfer to the fetus (40), and Boucher et al. showed glucuronidated BPA can induce markers of adipogenesis in 3T3L1 pre-adipocytes and primary human pre-adipocyte cultures (41). Overall, biomonitoring studies have shown that exposure to BPA is ubiquitous and according to Health Canada, BPA exposure has not substantially decreased since 2007 (42), emphasizing the need for mechanistic studies on its effects as an EDC and an obesogen.

1.2.3 BPA disrupts energy homeostasis

The obesogenic effects of BPA have become clear through human epidemiological studies, animal models of prenatal and postnatal BPA treatment and in vitro studies, illustrating the adipogenic and lipogenic potential of BPA (43, 44).

1.2.3.1 Humans

Multiple studies in Canada, the United States and Asia have reported that urinary BPA levels are positively associated with increased BMI, insulin resistance, diabetes and cardiovascular diseases (45-48). For example, in a Canadian Health Measures Survey conducted on 4733 adults, urine BPA concentration was associated with BMI with an odds ratio of 1.54 (46). In a US national health and nutrition examination survey (NHANES) study of 2104 adults, it was found that BPA levels positively associated with metabolic syndrome, which consists of increased blood pressure, diabetes, insulin resistance and obesity (49). Similar associations were found in reproductive-aged Korean women without metabolic syndrome: BPA levels were positively associated with BMI, waist circumference and HOMA-IR, a measure of insulin resistance (47). These associations are also seen in children and adolescents (48). Finally, a prospective study conducted in the US found no association between urinary BPA concentrations and baseline BMI; however, women with higher BPA concentrations had approximately 0.23 kg annual greater weight gain when followed for 10 years (50). These associations illustrate the obesogenic potential of BPA, but animal models were needed to imply causation.

8

1.2.3.2 Animal and in vitro models

Given the importance of the endocrine system during development, much of the research on BPA in animal models has exposed rodents to BPA in the prenatal, perinatal, or early postnatal timeframes. Though, a few studies investigating the effects of BPA on metabolic parameters after pubertal or adulthood exposure do exist. As mentioned above, the findings from these studies vary considerably as a result of the time of exposure and dose of BPA used per study. As such, when interpreting these studies, the possibility that those describing no effects at one or two doses may have missed the dose at which a response is seen cannot be ignored (14).

The United States Environmental Protection Agency (EPA) declared that the maximum tolerable daily intake (TDI) for BPA was <50 µg/kg bw/day, while the European Food Safety Agency has recently decreased this level to 4 µg/kg bw/day (14) in humans. Health Canada has set this limit at 25 µg/kg bw/day (51). Animal studies typically take these limits into account when dosing.

Rubin et al. reported increased body weight in male and female Sprague Dawley rats perinatally (gestational day 6 to lactation) exposed to 0.1 or 1.2 mg/kg bw/day from birth to 110 days of age (52). Female offspring exposed to the lower dose were heavier than those exposed to the higher dose, illustrating a non-monotonic response (52). Non-monotonic responses were also recorded in CD1 offspring that were exposed to 0.5 ng or 10 ng/kg bw/day from gestational day 15 to 18. Specifically, offspring exposed to the lower 0.5 ng/kg bw/day dose gained more weight. Furthermore, the increase in body weight was observed in female offspring at 12 weeks, which was not seen at 4 weeks, indicating a delayed response (53). Exposure of CD1 offspring to 5 to 50000 µg/kg bw/day BPA resulted in an increasing weight gain trend in only mice exposed to 500 µg/kg bw/day BPA, although glucose intolerance and insulin resistance was observed at lower doses (54, 55). The effects of perinatal BPA exposure were also found to be multigenerational as the F2 generation of exposed Wistar rats demonstrated higher body weights (56). High caloric diets can have additive effects with BPA as female CD1 mice perinatally exposed to BPA and subsequently fed a high fat diet (HFD) showed increased weight gain compared to those not exposed to BPA (57). This weight gain was accompanied by an increase in food intake and was not seen in male offspring (57). Nevertheless, several studies report decreased body weight or no change in body weight in response to BPA exposure (14). For example, maternal exposure of 50

9 ng, 50 µg or 50 mg BPA/kg diet resulted in increased energy expenditure, decreased food intake and decreased body weight in female offspring compared to controls (58).

Studies of short-term early postnatal exposure to BPA in rats and mice also have reported increased body weight in males and females. Remarkably, CD1 females exposed to 10 µg/kg bw/day from postnatal day 1 to 5 were 11% heavier than controls at 18 months of age (59). Finally, the obesogenic effect of adult-exposure to BPA has been described in animal models. Adult male and female C57BL/6J mice exposed to 5 to 5000 µg/kg bw/day BPA per day for 4 weeks showed increased body weight when fed a control diet, with accompanying increases in genes related to adipogenesis, fat synthesis and inflammation (60). Interestingly, the increased body weight occurred even at the lowest dose of BPA administered, much lower than the TDI of BPA of 50 µg/kg/day (14). Rubin et al. found 2.5 and 25 µg/kg bw/day BPA administered from 3 to 5 weeks of age exacerbates the effects of perinatal BPA exposure-induced weight gain in female, but not male CD1 mice (61). Furthermore, Moghaddam et al. showed that mice intraperitoneally exposed to 0.5 mg/kg/day BPA had a greater increase in body weight compared to control mice in as little as 14 days (62). Eight days of exposure to 100 µg/kg bw/day BPA did reduce both food intake and locomotor activity in mice; however, this was attributed to lethargy and was not associated with changes in weight (63). Nevertheless, these animals demonstrated impaired insulin sensitivity (63).

Thus, despite differences in doses and timing of exposure, there exists ample evidence of BPA- induced weight gain. In vitro models have clarified some of mechanistic aspects of these effects. For instance, BPA induces differentiation and lipid accumulation in murine and human preadipocytes through activation of peroxisome proliferator-activated receptor γ (PPARγ), glucocorticoid receptors (GR) or estrogen receptors (ER) (43, 44, 64), or through increasing levels of adipogenic enzymes, such as 11ß HSD type I (65). BPA also has been shown to increase insulin secretion from ß cells (66, 67). However, whether or not the hypothalamic control of energy balance is disrupted by BPA is less clear. Evaluating the response of the hypothalamic NPY/AgRP system, which coordinates food intake and energy expenditure, may shed light on whether hypothalamic disruption underlies any changes in body weight and glucose homeostasis seen in these animals. Some evidence of BPA-induced dysregulation of the NPY/AgRP system exists and is discussed below (Chapter 1.5).

10

1.2.4 BPA induces reproductive dysfunction

The impact of BPA on the reproductive system has been more broadly studied, as the estrogenic activities of BPA are well known. Epidemiological evidence suggests serum levels of BPA are positively associated with infertility and recurrent miscarriage, and inversely associated with peak E2 levels, quantity of oocytes retrieved, fertilization and implantation rates and embryo quality in women undergoing IVF treatments (68). Although the studies are more inconclusive, there is some evidence that increased BPA concentrations are associated with polycystic ovarian syndrome (69), endometriosis, uterine fibroids and preterm birth (14). Animal models have shown that BPA can impair ovarian development and hormone production (70), as well as uterine morphology (59). In addition, female Sprague-Dawley rats exposed to BPA perinatally had altered estrous cycles and decreased luteinizing hormone (LH) (52). In terms of BPA- induced male infertility, the primary evidence comes from occupational exposure to BPA leading to decreased sperm count, viability and motility (71). This holds true in the non-occupationally exposed population as well (72). In vitro studies suggest increased oxidative stress and DNA damage caused by BPA may underlie decreased sperm function (73, 74).

At the apex of the HPG axis and located in the hypothalamus, gonadotrophin releasing hormone (GnRH) neurons and kisspeptin (KISS) neurons regulate the reproductive axis and, thus, hormone production from the gonads (75). BPA has been shown to alter the activity of these neurons and/or the expression, secretion or function of the secreted by these neurons, as discussed below.

1.3 Hypothalamic control of energy homeostasis and reproduction

1.3.1 Energy homeostasis is controlled by NPY/AgRP and POMC neurons

The hypothalamus is the central region of the brain that controls energy homeostasis. Composed of different regions, called nuclei, the hypothalamus contains a heterogeneous population of neurons that respond to satiety signals, hunger signals and nutrients from the periphery and produce and that appropriately regulate food intake and energy expenditure, thus energy homeostasis and ultimately body weight (76). Hypothalamic nuclei that are involved in satiety and hunger were initially discovered using lesioning studies, and the lateral hypothalamus was defined as the hunger centre and ventromedial nucleus as the satiety

11 centre (77, 78). However, other nuclei, including the paraventricular nucleus (PVN), the arcuate nucleus (ARC) and the dorsomedial nucleus are also essential in the regulation of food intake (79). The ARC consists of two opposing populations of neurons, the orexigenic NPY/AgRP neurons and the anorexigenic POMC neurons (Figure 1-2). Due to the close proximity of the ARC to the median eminence (ME), which comprises a semi-permeable or “leaky” blood brain barrier, neurons in the ARC are able to sense circulating hormones and nutrients, including E2, leptin, insulin, fatty acids and glucose (76, 80), as well any circulating EDCs. In fact, BPA has been detected in human post-mortem hypothalamic tissue (37).

The POMC neurons of the ARC are stimulated by satiety signals and release α-melanocyte- stimulating hormone (α-MSH), which activates melanocortin receptors (MC3 and MC4) in the PVN to decrease food intake and increase energy expenditure. On the contrary, NPY/AgRP neurons are activated by the absence of such signals or by hunger signals, and secrete NPY and AgRP to increase food intake and decrease energy expenditure. NPY acts through NPY receptors

(Y1 to Y6), with Y1 and Y5 being the primary mediators of its orexigenic effects, while AgRP is an antagonist of the melanocortin receptors and partially acts by inhibiting α-MSH (80). This intricate balance between orexigenic and anorexigenic neurons can be dysregulated by exogenous factors, such as high fats and EDCs. Exogenous factors can aberrantly and directly increase NPY/AgRP and/or decrease POMC, overriding existing satiety and/or hunger signals, and can lead to hyperphagia resulting in obesity.

1.3.1.1 NPY and AgRP are potent orexigens

Abnormal increases in either NPY or AgRP are of particular concern as they are both potent orexigens, and changes in these alone are sufficient to alter whole body energy homeostasis. Intraceberoventricular (ICV) injection of either peptide acutely stimulates food intake and chronic administration leads to hyperphagia and ultimately obesity (81, 82). NPY and AgRP are increased following periods of energy deficiency (83, 84), and an inhibition of NPY protects against weight gain (82). Even modest overexpression of NPY (3.6-fold) is sufficient to induce obesity in adult rats (85) and a single dose of AgRP was shown to increase food intake for seven days (86). These sustained effects of AgRP are likely independent of the action of AgRP on melanocortin receptors as administration of an MC4 agonist 24 hours after AgRP administration did not prevent the increase in food intake (86).

12

1.3.1.2 NPY and AgRP regulate metabolism independent of food intake

NPY/AgRP neurons also regulate energy homeostasis and metabolism alongside or independent of food intake by projecting to distinct hypothalamic and extra-hypothalamic brain regions. Activation of AgRP neurons projecting to the lateral hypothalamus stimulates feeding and insulin resistance through modulation of brown adipose tissue (BAT) metabolism (87), whereas those projecting to the bed nucleus of the stria terminalis stimulate glucose uptake selectively into BAT (88). NPY has been shown to decrease heat production in BAT and increase lipogenesis and decrease lipolysis in white adipose tissue through inhibition of the sympathetic nervous system, primarily via the Y2 receptor (89). Furthermore, NPY administration to the brain increased fat deposition in the liver and white adipose tissue due to fatty acid and triglyceride synthesis (80, 89-91). AgRP itself is also thought to act independently of food intake as a derivative lacking the melanocortin receptor binding C-terminal region of AgRP was able to increase body weight and epididymal fat mass in the absence of increased food intake (92).

Therefore, NPY/AgRP neurons are critical regulators of metabolism and whether the obesogenic effects of BPA involve direct actions on NPY/AgRP neurons warrants investigation.

13

Figure 1-2 Hypothalamic control of energy homeostasis In the ARC of the hypothalamus, NPY/AgRP neurons and POMC neurons integrate hunger (ghrelin) and satiety (insulin and leptin) signals to regulate energy balance. α-MSH secreted from POMC neurons acts on MC4 on PVN neurons to suppress food intake and increase energy expenditure. In turn, NPY acts on Y1 or Y5 and AgRP antagonizes MC4 on PVN neurons to increase food intake and decrease energy expenditure. NPY/AgRP neurons and POMC neurons can reciprocally inhibit each other by activating Y1 or MC3, respectively. Exogenous factors, including EDCs, may disrupt this balance, leading to hyperphagia and obesity.

1.3.2 Reproduction is controlled by GnRH neurons

The hypothalamus coordinates reproductive function via the HPG axis. Briefly, GnRH neurons located in the medial pre-optic area of the hypothalamus secrete GnRH into the hypophyseal- portal system to induce secretion of follicle stimulating hormone (FSH) and LH from the anterior pituitary (75). FSH and LH travel to the gonads to stimulate production of E2, and testosterone as well as ovulation. Testosterone and E2 circulate back to the pituitary and hypothalamus to provide negative feedback (93).

14

1.3.2.1 Afferent regulators of GnRH neurons

GnRH neurons receive input from KISS and NPY neurons in the hypothalamus (Figure 1-3). Essential for GnRH secretion, KISS neurons exist as two subpopulations in the hypothalamus – the ARC kisspeptin neurons regulate pulsatile GnRH secretion, while the anteroventral periventricular (AVPV) nuclei KISS neurons regulate the pre-ovulatory surge in GnRH and LH (94). Other afferent regulators are thought to provide a mechanism for an additional layer of regulation by peripheral hormones and nutrients that indicate whether the appropriate nutritional status for reproduction has been met (75).

Evidence of distinct subpopulations of NPY neurons has been described in vivo and in vitro (95- 97). As such, subpopulations of NPY neurons regulate the HPG axis at the level of the GnRH neuron. In fact, immunoreactive NPY fibers have been visualized close to GnRH cells in the preoptic nucleus (98, 99). ICV injection of NPY blocks luteinizing hormone (LH) secretion in rats (100) and ewes (101), and food deprivation, which increases NPY levels, is associated with decreased LH secretion (102), illustrating the inhibitory actions of NPY on the reproductive axis. However, NPY also stimulates GnRH secretion (103-105); its levels increase prior to ovulation (106); and the LH surge, which triggers ovulation, is attenuated in animals lacking

NPY (107, 108). Furthermore, NPY increased GnRH mRNA in GT1-7 cells via Y1 activation. These differing effects of NPY are likely a result of activation of different receptor subtypes (109) as well as the hormonal milieu. E2-priming of ovariectomized rats allowed NPY- mediated increases in GnRH release, whereas in the absence of estradiol, NPY suppressed the reproductive axis (110). This finding is supported by the fact that E2 regulates NPY in vivo and in vitro (95, 111).

Padilla et al. recently demonstrated that AgRP neurons have inhibitory synaptic connections with KISS neurons in slices, and chemogenetic activation of AgRP neurons led to delayed estrous cycles and decreased fertility (112). Although γ-amino butyric acid (GABA) was shown to mediate this inhibitory action on KISS neurons, the potential role of NPY and AgRP was not ruled out. These studies suggested that in states of undernutrition, AgRP neuron activation may promote positive energy balance as well as supress reproduction in order to conserve energy (112). Thus, NPY/AgRP neurons are implicated in regulating reproductive function. How

15

BPA, which leads to reproductive dysfunction, perturbs this neuronal population requires further investigation.

KISS1R AR BA GA NPY KISS (AVPV or ARC)

HYPOTHALAMUS inhibitory

repress activate stimulatory + E2 KISS1 NPY Y 1 S1R KIS GABA Y 4 GnRH

Figure 1-3 Afferent regulation of reproductive neurons by NPY neurons NPY can stimulate or inhibit GnRH neurons depending on hormonal milieu and the NPY receptor subtype activated. NPY stimulates GnRH expression in the presence of E2 or via activation of Y4. In the absence of E2, NPY represses GnRH expression through Y1 activation. NPY neurons also inhibit KISS neuron activation through GABAergic projections. In turn, KISS neurons can increase NPY secretion from hypothalamic neurons through activation of KISS1R. Thus, NPY neurons play a role in the regulation of reproductive function.

1.3.3 Regulation of Npy and Agrp transcription and secretion

Npy and Agrp are regulated by hormones and nutrients, including E2, leptin, insulin, glucose, fatty acids and amino acids. Although the effects of these hormones and nutrients are established in vivo, the research regarding the specific mechanisms by which they regulate Npy and Agrp transcription or peptide secretion utilize cell models (see Table 1-1 for cell line information in Chapter 1.7) and in vitro molecular biology approaches (79). These studies have delineated some

16 of the signal transduction molecules as well as the transcription factors and promoter regions necessary for transcriptional regulation or secretion of the neuropeptides (Figure 1-4).

1.3.3.1 Regulation of Npy

E2 is an anorexigenic signal and has been shown in vivo to mediate its actions through NPY neurons (113, 114). E2 also acts as a negative regulator of the HPG axis, as well as a positive regulator to induce the pre-ovulatory LH surge (105, 107). As a result, E2 has population- specific effects on NPY neurons, downregulating Npy in certain populations, and upregulating Npy in others in vitro. Specifically, in mHypoE-42 cells, E2 downregulated Npy expression consistently over 72 hours through ERα activation, whereas in mHypoE-38 neurons, Npy was initially downregulated (~ 8 h) and required ERα and ERß, and subsequently upregulated at ~24 h strictly due to ERß activity (95). The increase in Npy in the mHypoE-38 neurons was accompanied by an increase in the ERß/ERα mRNA and protein ratio, corresponding to the dependence of the upregulation on ERß (95). Subsequently, it was found that the effects of E2 to downregulate Npy occurred through the action of E2 on membrane-bound ERα and activation of the phosphatidyl 3-kinase (PI3K)/Akt and extracellular signal-regulated kinase 1/2 (ERK1/2) mitogen-activated protein kinase (MAPK) pathways. These pathways converge onto the transcription factor cAMP response element-binding protein (CREB), which shows decreased binding to the proximal Npy promoter with E2, thereby decreasing transcription (115). The E2- mediated repression of Npy transcription was mapped to within -97 base pairs (bp) of the human Npy regulatory region (95). A longer putative repressor region, mapped to -1078 bp of the human Npy regulatory region, also binds transcription factors OCT-1 and PBX-1, which may be important for Npy repression in GT1-7 cells (116). The anorexigenic effects of E2 to decrease NPY secretion was also demonstrated to require membrane-bound ERα and the PI3K or 5’ adenosine monophosphate-activated protein kinase (AMPK) pathways (117).

In vivo, evidence of population specific effects of E2 on Npy-expressing neurons stems from the observation that only 10 – 20% of ARC NPY-immunopositive neurons in rats demonstrated estradiol accumulation within the nucleus (118). Furthermore, Esr1 expression in ARC NPY neurons is variable, suggesting that only certain populations are responsive to E2 or that ERα- independent actions of E2 occur in subpopulations of NPY neurons (119). Lastly, E2-mediated dampening of neuronal excitability occurs in part due to M-current generation via increasing

17 expression of KCNQ5 potassium channels. However, only 48% of NPY neurons expressed KCNQ5, again illustrating subpopulation-specific actions of E2 (120).

Leptin is an adipose-derived anorexigenic hormone that decreases Npy synthesis in the hypothalamus, but also modulates reproductive function (121, 122). As such, leptin was shown to regulate NPY secretion differentially in two hypothalamic cell lines. In the mHypoE-46 cells, leptin inhibited NPY secretion via a mechanism involving AMPK, whereas in the mHypoE-38 neurons, leptin upregulated NPY secretion and involved the MAPK and PI3K pathways (123). Interestingly, knocking out NPY in leptin-deficient ob/ob mice diminished the hyperphagia, obesity and infertility in ob/ob mice (124). The Y4 receptor is thought to play a role in increased NPY-mediated perturbations in leptin-deficient mice (125). Prolonged leptin treatment can induce leptin resistance, whereby the leptin-induced repression in NPY secretion is lost in vitro (126). Normally, leptin downregulates AMPK phosphorylation and activation, which is also blocked with prolonged leptin pre-treatment. Inhibiting AMPK or PI3K also prevented the leptin-mediated decrease in NPY secretion (126), suggesting that leptin resistance may occur as a result of the inability to downregulate AMPK or PI3K activity. AMPK is an energy-sensing molecule that is activated in low energy states. Indeed, activation of AMPK with 5- Aminoimidazole-4-carboxamide ribonucleotide (AICAR) increases NPY secretion by 30% in vitro (126) and constitutively active AMPK has been shown to increase the fasting-induced expression of both Npy and Agrp in the ARC (127).

As alluded to above, the cell line-specific regulation of Npy by E2 and leptin points to the existence of distinct subpopulations of Npy-expressing neurons. In fact, single cell RNA- sequencing studies of the murine hypothalamus have revealed distinct clusters of Npy-expressing neurons that may represent functionally distinct subtypes (128). Specifically, out of 34 neuronal clusters identified in the adult mouse hypothalamus, five clusters expressed Npy. Of these, only one cluster demonstrated co-expression of Npy and Agrp, and Agrp was only expressed in this one cluster. Of the five clusters expressing Npy, three were GABAergic and two were glutamatergic neurons, each with specific expression profiles of receptors and genes, illustrating the existence of distinct Npy-expressing neurons. Furthermore, in the three clusters with the highest expression of Npy, food deprivation for 24 hours led to distinct changes in their gene expression profiles, suggesting the potential for distinct functional events downstream of each

18 cluster (128). Other groups conducting single cell RNA-sequencing of the hypothalamus also report the extreme heterogeneity of hypothalamic neurons in vivo (129, 130).

Insulin is also an anorexigenic hormone and inhibits Npy and Agrp gene expression in vivo and in vitro (80, 131). In the mHypoE-46 cells, insulin decreased Npy and Agrp expression through a MEK-ERK-dependent pathway (131). Acting through the PI3K/Akt pathway, both leptin and insulin lead to phosphorylation of Forkhead box protein O1 (FOXO1). pFOXO1 is removed from the nucleus, thereby inhibiting Npy and Agrp transcription (132, 133). Thus, FOXO1 is a positive regulator of Npy and Agrp transcription. Signal transducer and activator of transcription 3 (STAT3) opposes FOXO1 and inhibits Npy and Agrp transcription (133, 134), although STAT3 alone is insufficient to induce leptin-mediated downregulation of the neuropeptides (135).

Palmitate is a saturated fatty acid, known to activate multiple pathways, including MAP kinases, toll-like receptor 4 and inflammation (136). With respect to Npy, palmitate increases Npy expression via AMPK activation in the mHypoE-44 cells (137) and/or through the nuclear factor κB (NFκB) pathway in the mHypoE-46 cells as inhibition of inhibitor of kappa B kinase (IκB kinase, IKK) (using PS1145) diminished the effects of palmitate on Npy (138).

1.3.3.2 Regulation of Agrp

The mechanisms by which E2 and insulin regulate Agrp expression in hypothalamic mHypoE-42 and mHypoE-38 or mHypoE-46 cells, respectively, are the same as Npy regulation. Estrogen downregulates Agrp in the mHypoE-42 cells, whereas it biphasically regulates Agrp in the mHypoE-38 cells (similar to Npy) (95). 10 nM insulin downregulates Agrp in the mHypoE-46 cells consistently from 2 to 12 hours (131).

Glucose-sensing AgRP neurons in the hypothalamus respond to increasing glucose concentrations by decreasing Agrp mRNA and secretion (139). Increasing glucose also decreases AMPK activation and increases pAkt, phosphorylated neuronal nitric oxide synthase (nNOS) and Ca2+/calmodulin-dependent protein kinase beta (CAMMKß) levels. In low glucose states, AMPK closes the cystic fibrosis transmembrane consductance regulator (CFTR), allowing for AgRP secretion. Increased glucose leads to activation of Akt, which phosphorylates nNOS, producing nitric oxide (NO). NO binds to soluble guanylyl cyclase, increases cyclic GMP levels, which

19 ultimately opens the CFTR and blocks AgRP secretion (139). In response to low glucose, activating transcription factor 3 (ATF3), which can be activated by AMPK (140), dimerizes with FOXO1 on the Agrp promoter, leading to increased transcription (141). The mammalian target of rapamycin (mTOR) pathway acts opposite to AMPK. Interestingly, amino acids inhibit Agrp gene expression through activation of the mTOR pathway (142), whereas the glucocorticoid dexamethasone upregulates Agrp via GR binding two glucocorticoid response elements in the Agrp 5’ regulatory region (143, 144). Again, these results demonstrate the direct relationship between AMPK activation and Agrp expression/secretion.

Thus, PI3K, MAPK and AMPK are important signaling proteins in the regulation of Npy and Agrp. ERs, CREB, FOXO1, STAT3, and ATF3 are transcription factors that bind to Npy or Agrp promoters to regulate their response to peripheral hormones and nutrients. As BPA can directly bind to ERs as well as phosphorylate and modulate signaling pathways (described below), it is plausible that the signaling molecules and transcription factors responsible for Npy and Agrp regulation are affected by BPA. Important to note, the regulatory region of mouse Agrp appears to lie up to 43 kb upstream of its transcriptional start site, although relatively few studies have investigated the Agrp 5’ regulatory region (145). Therefore, several other transcription factors currently not linked to Agrp may regulate Agrp transcription in response to BPA.

20

Figure 1-4 Regulation of NPY and AgRP expression and secretion (A) E2 and leptin both positively and negatively regulate Npy expression and/or secretion. Insulin suppresses Npy expression through decreased FOXO1 binding to the Npy promoter, whereas palmitate upregulates Npy partially through AMPK, NFκB and BMAL1 activation. (B) E2 and insulin use similar mechanisms to alter Npy and Agrp expression. Nutrients, including amino acid and glucose inhibit Agrp expression, whereas lowering glucose levels increases Agrp transcription by increasing FOXO1 and ATF3 binding to the Agrp promoter. Glucocorticoid dexamethasone also upregulates Agrp expression through increased GR binding to the Agrp promoter.

21

1.4 Hypothalamic control of circadian rhythms

Circadian rhythms are 24-hour patterns of activity that allow our bodies to adapt to the outside environment. Most biological functions that are regulated by the hypothalamus, including reproduction, feeding, stress, temperature regulation, blood pressure, sleep, activity and hormone production, exhibit circadian rhythms (146). These rhythms occur endogenously, but are entrained or synchronized by external zeitgebers (German for time givers), which include food, body temperature, social cues, and most importantly, light (147). These allow the internal clock to be synchronized with the external environment. Circadian rhythms are controlled by the suprachiasmatic nucleus (SCN), known as the ‘master clock’ due to its ability to entrain the rhythms of all other cells in the body. Located above the optic chiasm, the SCN receives input from light-sensitive retinal ganglion neurons through the retinal hypothalamic tract, and projects to various hypothalamic nuclei (i.e. medial preoptic area, PVN, dorsomedial nucleus) and brain regions to generate appropriate rhythms (146, 148).

Epidemiological studies and animal models both suggest that circadian disruption is linked to abnormal hypothalamic function. Shift workers, with disrupted circadian rhythms, experience reproductive dysfunction, with an increased risk of endometriosis and miscarriage rates and irregular menstrual cycles (149, 150), as well as increased rates of diabetes, metabolic syndrome and obesity (151, 152). Likewise, animals exposed to constant light had dampened rhythms in insulin sensitivity (153). SCN lesions disrupt the LH surge and estrous cycle, increase body weight and ablate rhythms in food intake and locomotor activity in rodent models (154-156), and transplantation of SCN grafts into lesioned animals restores many of these rhythms (157, 158).

Mechanistically, these rhythms are controlled by the clock genes, participating in a transcriptional-translational feedback loop (Figure 1-5). Brain and muscle ARNT-like 1 (Bmal1) and circadian locomoter output cycles protein kaput (Clock) are transcribed, translated, and dimerize to act as transcriptional regulators of period (Per1-3), cryptochrome (Cry1-2), nuclear receptor subfamily 1, group D, member 1 (Rev-erbs), retinoic acid-related orphan receptor (RORs) and Ppars by binding to E-box elements in their promoters. In turn, PER and CRY translocate into the nucleus to interact with the BMAL1:CLOCK heterodimer to inhibit their own transcription, while REV-ERBs, RORs and PPARs modulate the transcription of Bmal1. Together with their intrinsic rates of degradation, this feedback loop allows Per and Cry to

22 oscillate over a 24 h period with Bmal1 expression changing in an antiphasic manner. The 24 h period is maintained by post-translational modifications of PER and CRY that delay their feedback onto the BMAL1:CLOCK heterodimer (147). In addition to regulating the transcription of clock-related genes, BMAL1 and CLOCK are transcription factors that can regulate the expression of several E-box containing genes (147). As a result, approximately 43% of mammalian protein coding genes and many non-coding genes show rhythmicity (159). These clock genes are tightly linked to energy and reproductive homeostasis as mutations in them lead to pathogenesis. For instance, Clock and Per2 mutants develop obesity (160, 161), while Clock and Bmal1 mutant mice have impaired LH surges and estrous cycles (162, 163). In humans, polymorphisms in many of the clock genes have been associated with metabolic syndrome and reproductive dysfunction (164).

ROR repress increase PPAR

Per, Cry, Rev-erb, CLOCK BMAL1 Ror, Ppar E-box Gene

REV- ERB

PER CRY

Figure 1-5 Transcriptional-translational feedback loop underlying generation of circadian rhythms|

Bmal1 and Clock are transcribed, translated and bind to E-box elements to upregulate the expression of Per1-3, Cry1-2, Reverbs, Rors and Ppars. Upon translation, PER, CRY and REV- ERBs negatively regulate BMAL1 expression and/or activity thereby inhibiting their own transcription, while RORs and PPARs positively regulate Bmal1 transcription. This feedback mechanism allows for the generation of a 24 h period.

23

1.4.1 Circadian involvement in Npy regulation

Obesity-promoting factors, such as a HFD or more specifically palmitate, can dysregulate the expression of clock genes and these disruptions can underlie changes in Npy expression. For example, palmitate increased Bmal1 expression and decreased Per2 expression in mHypoE-44 and mHypoE-37 cells (137, 165). The amplitude of rhythmic Npy expression in the mHypoE-44 cells was also increased with palmitate treatment (166). This increase was attributed to the activation of AMPK by palmitate (166); however, the involvement of BMAL1 in this process was suspected as BMAL1 directly binds the Npy promoter (167). Since then, using the mHypoA- BMAL1-KO/F cell line, lacking BMAL1, alongside its wildtype control, it was found that palmitate was unable to induce Npy expression in mHypoA-BMAL1-KO/F cells (168) Thus, the circadian clock in hypothalamic neurons can be altered by exogenous compounds, contributing to hypothalamic neuropeptide dysregulation. Particularly, Npy expression is vulnerable to circadian dysregulation.

Perinatal exposure of the California mouse strain to 50 mg/kg BPA in diet abolished diurnal food intake patterns in male offspring, and female offspring consumed more food in the light phase as opposed to the dark phase (169), indicating disruption in the circadian regulation of food intake induced by BPA. Whether BPA induces dysregulation of circadian genes in NPY/AgRP neurons that control food intake is unknown.

1.5 Effects of BPA in the hypothalamus

The fact that BPA affects hypothalamic neurons is well known, but the consensus on how exactly it is affected is not clear. This is again not surprising with inconsistencies in the doses administered, time of endpoint analysis and region of hypothalamus examined (14). Regardless, prenatal BPA exposure has been shown to alter brain steroid receptor levels (170), alter sexually dimorphic areas of the brain (171), and lead to behaviours such as hyperactivity and anxiety (172).

Firstly, ER levels in the hypothalamus in response to BPA have been documented in several studies. ERα mRNA and protein levels were generally upregulated across studies, however, there were region specific differences (14). For instance, postnatal exposure to BPA upregulated ERα in the AVPV and decreased it in the ARC at postnatal day 100 (170). Yet again, other studies

24 showed no changes in ERα levels in the ARC (14, 173). Similar variability is seen with ERß, progesterone receptor and glucocorticoid receptor in prenatally exposed mice or rats (14, 170, 174). Kundacovic et al. demonstrated sex- and region- specific alterations in ERα and concomitant changes in DNA methyltransferase 1 and 3 in the cortex of male mice and the hypothalamus of female mice leading to sex-specific changes in anxiety-like behaviour (172). These studies illustrate the importance of examining specific brain regions, as use of the whole brain can mask region-specific effects (14). Aromatase protein or mRNA levels were upregulated in the hippocampus and prefrontal cortex of rodents exposed to BPA (175, 176). Furthermore, in a zebrafish model, BPA exposure resulted in a 180% increase in neurogenesis in the hypothalamus, which was mediated by BPA activating the androgen receptor (AR) and leading to the upregulation of aromatase (177).

BPA influences the hypothalamic control of reproduction by directly altering GnRH and KISS neurons. In female rats, neonatal exposure to low doses of BPA delayed the developmental decrease in the GnRH interpulse interval that is essential for sexual maturation. However, higher doses led to an early decrease in this pulsatility pattern resulting in early puberty (178). In males exposed to low doses of BPA from gestational day 18 to postnatal day 5, hypothalamic GnRH expression was increased and accompanied by delayed puberty (179). KISS neuron immunoreactivity was also decreased in the ARC with high doses of BPA (180), and 10 nM of BPA directed into the stalk median eminence of monkeys decreased the levels of GnRH and kisspeptin 1 (KISS1) released (181). In vitro, BPA has been shown to directly decrease the firing activity of GnRH neurons (182). Whether NPY/AgRP neurons contribute to the aberrant regulation of GnRH and KISS1 by BPA remains unknown.

The effects of BPA on the hypothalamic control of energy balance are beginning to be studied with prenatal/early postnatal exposure, although mechanistic studies are still lacking. Mackay et al. showed Npy and Agrp mRNA levels were increased after perinatal BPA exposure (~ 3.49 µg/kg day) in male mice subsequently fed a HFD. These BPA-exposed animals also displayed decreased POMC fiber innervation in the PVN and had impaired glucose tolerance, even in the absence of the HFD (57). Female offspring exposed to BPA from the same study displayed increased weight gain and food intake, and decreased Pomc mRNA levels in the ARC, when subsequently fed a HFD (57). A later study by the same group showed that perinatal exposure to BPA prevented leptin-induced reduction in food intake and body weight loss, and in leptin-

25 mediated POMC upregulation (183). More recently, Desai et al. illustrated perinatal BPA exposure in mice increased AgRP and decreased POMC protein levels in neural progenitor cells of newborn mice. Furthermore, neural progenitor cells from control mice treated in vitro with 10 µM BPA for 5 days showed increased levels of NPY and AgRP and decreased POMC protein as measured by Western Blot (184). In addition, neonatal rats exposed intracisternally to BPA for 5 days demonstrated slight increases in Npy mRNA in the midbrain, but not in the striatum (185). These studies clearly illustrate the orexigenic potential of BPA at the level of the hypothalamus; however, the molecular mechanisms underlying these effects are unknown. Studying the BPA- mediated regulation of neuropeptides at a more defined mechanistic level is crucial to design therapeutic strategies to prevent the detrimental effects of BPA, as well as to enhance the information available for regulatory bodies that limit the use of EDCs in society.

It is important to note that two studies report null or opposite effects to those seen above in terms of Npy and Agrp expression. Fischer CDF female rats exposed to 50 mg/kg/day or 50 µg/kg/day BPA from embryonic day 18 to postnatal day 7 did not show any changes in hypothalamic Agrp, Esr1, Esr2 or Foxo1 expression at adulthood (186). These animals also did not display altered timing of puberty or body weight (186). Second, zebrafish exposed to 4 or 4000 µg/kg/day for 21 days had decreased Npy expression (187). These differential effects may be a result of species- or strain-specific effects, BPA dosage, timing of exposure or timing of endpoint measurement. In support of this possibility, Yang et al. reported more pronounced increases in body weight and fat mass in male and female C57BL/6J mice exposed to 5 µg/kg/day and 500 µg/kg/day BPA compared with an intermediate dose of 50 µg/kg/day (60), which was the dose used in the Fischer CDF rat study mentioned above (186).

1.6 Signaling pathways activated by BPA

The mechanisms by which BPA elicits a specific effect are difficult to study because it binds to an assortment of receptors and activates multiple signaling pathways (Figure 1-6). Thus, the potency of the effect does not depend on a single interaction and the number of possible interactions may differ with the dose of BPA used (14).

26

1.6.1 Hormone and nuclear receptors

Initial studies considered BPA an estrogen mimic with approximately 1000-fold lower affinity, as it was shown to activate ERs in several in vitro and in vivo experiments (188). However, nuclear receptor estrogenic effects and rapid effects through binding membrane-associated G- protein-coupled ER (GPER) were demonstrated even at lower doses (189). Whether BPA acts as an agonist to the ERs is debated, however, as the conformational change in the receptors and the specific set of co-regulators that are recruited by BPA upon ER binding are different from that of E2 (190, 191). Depending on the tissue type and the concentration of BPA, several other nuclear receptor and endocrine pathways are shown to mediate BPA-induced effects. Alternative mechanisms include activation of ERRγ, aryl hydrocarbon receptor (AhR), PPARγ, GR (64) or AR (177) and a correlation to increased levels of 11β-hydroxysteroid dehydrogenase type 1 (65).

ERRγ is an orphan nuclear receptor with no known endogenous ligand; however, analysis of the crystal structure of its ligand binding domain has revealed that it binds BPA at a much greater affinity compared to classic ERs (192). ERRγ is predominantly expressed in the placenta; however, it is expressed in the brain, where isoform 2 is the predominant form (193). Tohmé et al. demonstrated that mutations of the ERRγ ligand-binding domain, which reduced the ability of BPA to bind the receptor, diminished certain BPA-induced developmental abnormalities in zebrafish. The mechanism of action was linked to the ability of ERRγ to act as a transcriptional regulator (194). ERRγ can bind to the ERR consensus sequence as a monomer or to estrogen response elements to regulate gene expression (192).

AhR has been postulated to bind and mediate the toxic effects of several environmental chemicals, containing aromatic rings (195). A study published by Caserta et al. positively correlated serum BPA levels to increased expression of AhR in white blood cells of infertile women (196). Furthermore, low doses of BPA given to pregnant mice led to increased mRNA levels of AhR in the brains of male and female offspring, implicating potential involvement of the receptor. In fact, exposure to BPA was shown to activate AhR through luciferase reporter gene assays and lead to the binding of the AhR complex to specific DNA response elements in human hepatocytes (197). Ziv-Gal et al. identified that the inhibition of ovarian follicle growth by BPA was partially recovered in AhR knockout mice (198). AhR is a steroid receptor that translocates into the nucleus upon ligand binding and dimerizes with aryl receptor nuclear

27 translocator (ARNT). This complex can bind to xenobiotic response elements (XRE) and act as a transcription factor (195).

PPARs are a family of nuclear receptors, including PPARγ, PPARα and PPARδ, which have predominantly been linked to fatty acid metabolism in the periphery (199). However, these receptors are also expressed in the hypothalamus at relatively high levels. Rosiglitazone, a PPARγ agonist, increased the expression of AgRP in the arcuate nucleus of mice and Siberian hamsters (200). BPA induces adipocyte differentiation in 3T3-L1 cells and human adipocytes and this was dependent on PPARγ activation (43). Thus, BPA can bind to and activate PPARγ, leading to its translocation into the nucleus, where it can act as a transcription factor.

It is not known which of these receptors may bind BPA in hypothalamic neurons. The concentration of the chemical, the expression levels of receptors and the inherent nature of the cell types may lead to tissue-specific mechanisms.

1.6.2 Non-nuclear receptor pathways including induction of cellular stress

In addition to binding nuclear receptors, BPA has been linked to the activation of neuroinflammatory, endoplasmic reticulum stress and oxidative stress pathways. Neuroinflammation is characterized by the production of cytokines, such as tumor necrosis factor alpha (TNFα) and interleukins. These cytokines are typically produced in response to a pathogen and recruit glial cells to resolve the insult (201). However, exogenous factors, such as saturated fats and EDCs, can inadvertently activate the pathways that regulate the production of these cytokines, leading to neuroinflammation (202). These pathways include MAPKs and NFκB, which are activated by BPA (203, 204). Furthermore, in females, urinary concentrations of BPA were positively correlated to increased inflammatory markers, including TNFα levels (60).

Endoplasmic reticulum stress (EndR stress) is a consequence of an accumulation of unfolded proteins in the cell. This initiates what is known as the unfolded protein response (UPR). The UPR upregulates transcription factors that produce chaperone proteins to help refold proteins and shuts down general translation activity. These transcription factors include DNA damage inducible transcript 3 protein (DDIT3/CHOP), activating transcription factors (ATF2, 3, 4, 6)

28 and X-box binding protein 1 (XBP1), among others (205, 206). The EndR stress activating capabilities of BPA have been described in other systems (207, 208).

Oxidative stress results from excess production of reactive oxygen species (ROS) and nitrogen species that cannot be controlled by cellular antioxidant defenses. Oxidative stress can cause calcium influx from outside the cell as well as the endoplasmic reticulum, resulting in misfolded proteins, and EndR stress (209, 210). BPA has been shown to lead to oxidative stress by binding to and inactivating major antioxidant enzymes, including superoxide dismutase (SOD) and catalase (CAT) (73, 211, 212) and by altering calcium influx via interaction with ion channels (213-215). Oxidative stress markers are correlated with higher urine levels of BPA (216) and are increased in sperm exposed to BPA (212). This increase in oxidative stress can aberrantly activate signaling proteins, such as MAPKs and AMPK to ultimately alter gene expression (209, 217). All three forms of cellular stress occur in obesity (201, 206, 218).

Figure 1-6 Common signaling pathways activated by BPA (A) As an endocrine disrupting chemical, BPA binds to and activates several nuclear receptors (ERRγ, AhR, PPARγ, ER and GR) that act as transcription factors by binding to gene promoters. (B) As an estrogen mimic, BPA also binds to membrane bound estrogen receptor, GPER, which resides on both the plasma membrane or the EndR membrane. Alternatively, BPA activates cellular stress responses, including (C) inflammation, (D) endoplasmic reticulum stress and (E) oxidative stress. Stress-induced activation of the unfolded protein response involving IRE1, PERK and ATF6 ultimately leads to the activation of transcription factors (ATFs and XBP1) and co-

29 regulators (CHOP), which can modulate gene expression. This resulting EndR stress response can lead to activation of MAP3Ks. MAP3Ks can promote an inflammatory response by allowing nuclear translocation and activation of NFκB. BPA induces oxidative stress by decreasing the activity of antioxidant enzymes, SOD1 and CAT. Oxidative stress can in turn lead to calcium influx from the extracellular environment as well as the EndR, increasing mitochondrial dysfunction and leading to further ROS production. The resulting build-up of ROS can increase MAPK (ERK and JNK) and AMPK phosphorylation, leading to signaling cascades responsible for Npy and Agrp transcription. 1.7 Hypothalamic cell models to study BPA-induced dysregulation

In order to address the question of whether hypothalamic Agrp and Npy neuropeptides may play a role in the obesity and reproductive dysfunction caused by BPA, this thesis focused on establishing whether Agrp and Npy are targeted by BPA at the transcriptional level. In the whole hypothalamus, the heterogeneous neuronal subtypes as well as the existence of glial cells renders it difficult to elucidate whether exogenous factors directly affect individual neuronal subtypes defined by the expression of specific neuropeptides. Although neuropeptide expression in subpopulations of neurons may be altered by BPA, these changes can be masked when the RNA from the whole tissue is pooled. As the hypothalamus consists of functionally distinct subpopulations of NPY/AgRP neurons that serve to differentially regulate afferent neurons, as previously discussed (Chapter 1.3), it is important to establish how these neurons respond to BPA on an individual level. Furthermore, how BPA may act to alter neuropeptides is difficult to study without the use of cell lines where receptors and pathways can be experimentally manipulated to delineate specific mechanisms of action.

Over the last 20 years, the Belsham laboratory generated immortalized cell lines representing clonal populations of adult and embryonic, male and female-derived hypothalamic neurons expressing and secreting NPY and AgRP. The specific cell lines used in this study are represented in Table 1-1 and previous studies have characterized these cell lines describing their appropriate response to peripheral signals, such as insulin, estrogen and leptin as described in Chapter 1.3.3) (Table 1-2). The availability of a wide array of cellular models in our laboratory allows for delineation of molecular mechanisms in specific heterogeneous neuronal populations.

30

Table 1-1 Characteristics of cell lines used in current study

Clonal or Embryonic Sex of cell Neuropeptide Cell Line mixed or adult line expression population mHypoE-41 Embryonic Female Clonal Npy , Agrp mHypoA-59 Adult Female Clonal Npy , Agrp mHypoE-46 Embryonic Male Clonal Npy , Agrp mHypoE-44 Embryonic Male Clonal Npy , Agrp mHypoE-42 Embryonic Male Clonal Npy , Agrp mHypoA-2/12 Adult Male Clonal Npy , Agrp mHypoA-BMAL1-WT/F Adult Female Mixed Npy , Agrp , Pomc mHypoA-BMAL1-KO/F Adult Female Mixed Npy , Agrp , Pomc mHypoA-BMAL1-WT/M Adult Male Mixed Npy , Agrp , Pomc mHypoA-BMAL1-KO/M Adult Male Mixed Npy , Agrp , Pomc m = mouse, E = embryonic, A = adult, WT = wildtype, KO = knockout, F = female, M = male

31

Table 1-2 Regulation of (A) Npy and (B) Agrp in cell lines used in current study

mHypoX-xx, where X=E or A and xx=cell line number ~ = no change, n.d. = not done

Primary cultures of hypothalamic neurons were also used to a limited extent to determine the overall response of a mixed population of non-immortalized hypothalamic neurons to BPA treatment. However, these were only used to determine initial responses at one time point and with one concentration as very few neurons are cultured from a single hypothalamus and neurons cannot be passaged without increased cell death. Furthermore, the unpredictable heterogeneity between culture plates renders mechanistic studies difficult, even if it were feasible.

Finally, in order to study the involvement of BMAL1 in the response to BPA, immortalized cell lines were derived from the hypothalamus of male and female BMAL1-KO mice and their

32 wildtype littermate controls: mHypoA-BMAL1-KO/F and /M and mHypoA-BMAL1-WT/F and /M. These represent a mixed population of hypothalamic neurons and each line expresses Npy, Agrp and Pomc. Although the expression of other clock genes, such as Per2, are intact in these cell lines, the rhythmicity in their expression is lost due to the lack of BMAL1 (168), making these lines appropriate tools to study neuroendocrine responses to circadian dysregulation. Involvement of hypothalamic specific BMAL1 in homeostatic and pathological responses to hormones, nutrients and chemicals can also be evaluated.

1.8 Hypothesis and Aims

As described, BPA leads to perturbations in many aspects of energy homeostasis, including food intake, energy expenditure, fat mass, body weight and glucose homeostasis. BPA also disrupts the HPG-axis and contributes to reproductive dysfunction that could lead to infertility. NPY/AgRP neurons orchestrate many of these physiological processes that begin at the level of the hypothalamus. As such, whether the NPY/AgRP neurons, and specifically Npy and Agrp expression, are affected by BPA and whether this underlies the BPA-induced metabolic and reproductive perturbations remains unknown.

Therefore, we hypothesize that BPA increases Agrp and Npy gene expression in hypothalamic neurons and that this effect is mediated through nuclear receptor activation, induction of cellular stress and subsequent transcription factor activation and/or circadian dysregulation. To investigate this hypothesis, we used six immortalized NPY/AgRP-expressing cell lines as well as cell lines lacking circadian protein BMAL1. These were used alongside pharmacological antagonists or siRNAs against nuclear receptors and transcription factors as well as inhibitors of the different forms cellular stress (inflammation, EndR stress, oxidative stress).

The experiments conducted to test the hypothesis are presented as the following three aims:

Aim 1: Determine if BPA can alter Agrp gene expression in different subpopulations of Agrp- expressing neurons and elucidate the signaling pathway or transcription factors responsible for these effects. These results are presented in Chapter 3.

33

Aim 2: Examine the effects of BPA on Npy gene expression in different subpopulations of Npy- expressing neurons and elucidate the underlying mechanisms. This aim will also investigate whether there are differential effects on the expression of estrogen receptors in the subpopulations of Npy neurons. These results are presented in Chapter 4.

Aim 3: Determine the effects of BPA on circadian gene expression in hypothalamic neurons and investigate whether circadian dysregulation underlies the changes in Npy and Agrp expression. These results are presented in Chapter 5.

Chapter 2 Materials and Methods

34 35

2 Materials and Methods 2.1 Cell culture and reagents

2.1.1 Immortalized cell lines

Neurons from adult (A) or embryonic (E) mouse hypothalamii (mHypo) were immortalized as previously described (219, 220) to generate several Npy-expressing clonal female-derived (mHypoA-59, mHypoE-41), and male-derived (mHypoA-2/12, mHypoE-42, mHypoE-46, mHypoE-44) cell lines (Figure 2-1). Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM; MilliporeSigma, Oakville, ON, Canada) with 4500 mg/L glucose, supplemented with 2% fetal bovine serum (FBS; Gibco, Burlington, ON, Canada) and 1% penicillin-streptomycin

(PS; Gibco) at 37 C with 5% CO2. Cells were split into 60 mm tissue culture dishes approximately 24 hours prior to treatment and grown to 75 – 85% confluency. Growth media was replaced with treatment media (as described below) on the day of treatment.

2.1.2 Primary culture

For primary culture experiments, cells from the hypothalami of 8-week old male or female CD-1 mice (Charles River Laboratories, Senneville, QC, Canada) were dispersed by trituration. Cells were cultured in 6-well plates in neurobasal A medium (Gibco), supplemented with 10% FBS, 5% horse serum (Gibco), 1% PS (Gibco), 1 x B27 serum-free supplement (Gibco) and 1 x GlutaMAX supplement (Gibco) for 7-9 days. Each well received 10 ng/L ciliary neurotrophic factor (CNTF, R&D Biosystems, Oakville, ON, Canada) once per day to induce proliferation. After 7-9 days, cells were treated with vehicle (0.05% EtOH) or 100 M BPA (as described below) for 8 hours in DMEM modified phenol red-free media, supplemented with 1% CSFBS (Gemini Bio Products) and 1% PS (Gibco). All animal procedures were conducted in accordance with the regulations of the Canadian Council on Animal Care and approved by the University of Toronto’s animal care committee.

2.1.3 Preparation of mHypoA-Bmal1-WT and mHypoA-Bmal1-KO cell lines

Bmal1 heterozygous mice (Stock #009100, The Jackson Laboratory, ME, United States) were purchased and bred by Dr. Patricia Brubaker’s laboratory to obtain Bmal1-knockout (KO) mice and wildtype (WT) littermates. The hypothalami of two Bmal1-KO mice and two wildtype (WT)

36 littermates were then obtained from Dr. Brubaker and neurons were immortalized as previously described (220). Briefly, the hypothalamus of an 8 - 9-week-old female Bmal1-KO mouse, male Bmal1-KO mouse, female littermate control, and male littermate control, were separately isolated and dispersed into primary culture. Primary cultures were treated with 10 ng/L CNTF for 5-7 days to induce neuronal proliferation, followed by viral transformation of a plasmid containing the SV40 T-antigen and a neomycin resistance cassette to induce immortalization. Immortalized cells were selected for using 100 g/ml G418 (Geneticin, Gibco). This generated four cell lines representing a mixed population of hypothalamic neurons: mHypoA-Bmal1-KO/F, mHypoA- Bmal1-KO/M, mHypoA-Bmal1-WT/F, and mHypoA-Bmal1-WT/M. Cell lines were screened for Bmal1 expression, circadian clock gene expression, neuropeptide expression and related markers using real-time quantitative PCR. Cells were grown to 70-75% confluency in DMEM containing 4500 mg/l glucose (MilliporeSigma), 2% FBS (Gibco) and 1% PS (Gibco) and treated with vehicle (0.05% EtOH) or 100 M BPA in DMEM modified phenol red free media (1% CSFBS, 1% PS) for 4 or 8 hours. Experiments in mHypoA-Bmal1-KO cell lines were conducted in parallel with mHypoA-Bmal1-WT cells for comparison of effects. All animal procedures were conducted in accordance with the regulations of the Canadian Council on Animal Care and approved by the University of Toronto’s animal care committee.

2.1.4 BPA treatment

BPA (MilliporeSigma) was dissolved in 100% EtOH to a concentration of 200 mM, then diluted

1:1 in sterile H2O to obtain a 100 mM stock solution. Vehicle (50% EtOH + 50% H2O) or 100 mM BPA was then diluted 1:1000 in phenol-red free DMEM containing 4500 mg/L glucose (Hyclone Laboratories Inc., Whitby, ON, Canada), supplemented with 1% charcoal-dextran stripped FBS (CSFBS; Gemini Bio Products, Burlington, ON, Canada) and 1% PS, giving a final concentration of 0.05% EtOH or 100 M BPA. For dose curve experiments, 200 mM BPA was further diluted in 100% EtOH to 100 mM, 50 mM or 20 mM BPA, then diluted 1:1 in H2O, followed by 1:1000 in media to obtain final concentrations of 50 M, 25 M and 10 M BPA in 0.05% EtOH. Cells were washed with phosphate-buffered saline (PBS) and then treated with BPA for 2 to 24 hours as detailed in figure legends.

37

Adult C57BL/6 Bmal1-WT mouse/ Embryonic or mouse* Bmal1-KO mouse* mouse

HYPOTHALAMIC EXTRACTION

DISPERSION (PRIMARY CULTURE) SV-40 T-Antigen CNTF*

Geneticin GROWTH

SUBCLONING

EMBRYONIC mHypoA-Bmal1-WT or CLONAL ADULT mHypoA-Bmal1-KO CELL LINES CLONAL CELL LINES

Figure 2-1 Generation of immortalized hypothalamic cell lines Following extraction and dispersion of the hypothalamii from embryonic, adult C57BL/6, Bmal1- WT or Bmal1-KO mice, cells were cultured and transformed with a vector containing SV-40 T- Ag and neomycin resistance gene. Immortalized cells were selected for using geneticin and subcloned to create individual populations of clonal embryonic and adult cell lines. For adult- derived cultures (*), cells were treated with CNTF to induce neuronal proliferation prior to immortalization. mHypoA-Bmal1-WT and mHypoA-Bmal1-KO cell lines were not further subcloned and represent a heterogeneous population of neurons.

38

2.2 BPA content enzyme-linked immunosorbent assay (ELISA)

Cells were grown to 75-80% confluency in 60 mm tissue culture plates 24 hours prior to treatment with 0.05% EtOH, 10 M BPA or 100 M BPA for 4 or 24 hours in phenol-red free DMEM, supplemented with 1% CS-FBS and 1% PS. After treatment, 500 L of media from each plate was collected, immediately dried down using a vacuum concentrator and stored at -20 °C. Cells were lysed with 500 L/plate 1 x cell lysis buffer (Cell Signaling Technology Inc., New England Biolabs, Whitby, ON, Canada) diluted in PBS. BPA was then collected from the cells using an ethyl acetate extraction method. Briefly, 250 L of cell lysate was acidified with 1 L of acetic acid to obtain a pH of 3-4. 250 L of ethyl acetate was added to each sample, vortexed and centrifuged at 12 000 x g for 3 minutes to separate organic and aqueous layers. The top organic layer was collected and the extraction was repeated twice using 250 L ethyl acetate each time. A total of 700 L organic layer was obtained after three extractions, which was immediately dried using compressed air and stored at -20 °C. To ensure efficient recovery of intracellular BPA with the ethyl acetate extraction method, empty culture plates were treated with 0.05% EtOH, 10 M BPA or 100 M BPA and the amount recovered in the media (with no cells) was compared to the amount recovered in the cells + media. These values were comparable, suggesting appropriate recovery of BPA from the cells. The Estrogen BPA Environmental ELISA Kit (cat. #ab175820) was purchased from Abcam Inc. (Toronto, ON, Canada) and the assay to measure BPA content was performed as per manufacturer’s instructions. A 1/50 or 1/100 dilution of each sample in sample dilution buffer was performed prior to loading samples into assay wells. Percent BPA content was calculated using the following formula: % BPA content = [(content in cell lysate)/(content in cell lysate + content in media)]*100.

2.3 Inhibitors, antagonists and supplements

For inhibitor pre-treatment experiments, pre-treatments were added to the cells in 2.5 mL media. BPA or EtOH was then added to the cells in 0.5 mL media to achieve final concentrations of 100 M BPA or 0.05% EtOH in 3 mL volume.

39

2.3.1 Inhibitors

Stock solutions of actinomycin D (ActD 10 mg/ml), SP600125 (50 mM), PD0352901 (10 mM), Compound C (CC 6 mM) and nordihydroguaiaretic acid (NDGA 4 mM) were prepared by dissolving in 100 % dimethyl sulfoxide (DMSO), followed by a 1:1000 dilution in media to obtain final concentrations of 10 g/ml ActD, 50 M SP600125, 10 M PD0352901, 6 M CC and 4 M NDGA in 0.1% DMSO. Sodium phenylbutyrate (PBA) and tauroursodeoxycholic acid

(TUDCA) were dissolved in sterile H2O to a concentration of 500 mM or 250 mM, respectively, followed by a 1:100 or 1:50 dilution into media to obtain final concentrations of 5 mM and 10 mM PBA and 2.5 mM TUDCA. PS1145 dihydrochloride (PS1145) was dissolved in 100% DMSO to 10 mM, and diluted 1:500 in media to a concentration of 20 M in 0.02% DMSO. ActD (cat. #A1410-5MG), PBA (cat. #SML0309-100MG), CC (cat #. 171260), TUDCA (cat #. 580549), NDGA (cat #. 74540) and PS1145 (cat. #P6624) were purchased from MilliporeSigma. SP600125 (cat. #1496/10) and PD0352901 (cat. #4192/10) were purchased from Tocris Bioscience (Cedarlane, Burlington, ON, Canada). Cells were pretreated with these compounds or vehicle for 1 hour, followed by 100 M BPA or 0.05% EtOH for 8 or 16 hours.

AICAR (cat #. 2840/50) was purchased from Tocris BioScience, dissolved in H2O to 50 mM, and diluted 1:100 in media for a final concentration of 500 M. Cells were pretreated with 500

M AICAR or H2O for 1 hour, followed by 0.05% EtOH treatment for 8 hours or 16 hours.

2.3.2 Steroid receptor antagonists

Stock solutions of all steroid hormone receptor antagonists were prepared by dissolving in 100% DMSO. These were then diluted 1:200 (for 5 M G15 only) or 1:1000 in media to obtain the final concentrations listed. The GPER antagonist G15 (cat. #3678), the ER antagonist PHTPP (cat. #2662) and the AhR antagonist CH223191 (cat. #3858) were purchased from Tocris BioScience (Cedarlane, Burlington, ON, Canada). The ERR inverse agonist GSK5182 (cat. #AOB1629) was purchased from Aobious Inc. (Gloucester, MA, USA). The GR antagonist RU486 (cat. #M8046) and the PPAR antagonist T0070907 (cat. #10026) were purchased from MilliporeSigma and Cayman chemicals (Cedarlane, Burlington, ON, Canada), respectively. Cells were pre-treated with antagonists or vehicle for 1 hour, followed by treatment with 100 M BPA or 0.05% EtOH for 8 or 16 hours.

40

Concentrations of steroid receptor antagonists were selected based on previous experiments and each antagonist was confirmed to block the appropriate receptor by measuring the response of a gene known to be targeted by the receptor. G15 and PHTPP have previously been used in our laboratory and have been shown to block estrogen-mediated effects (221). Furthermore, the BPA-mediated decrease in GPER-target estrogen-related receptor α (Esrrα) and ER-target insulin-like growth factor (Igf1) (222) were mitigated with G15 and PHTPP, respectively. The BPA-mediated decrease in Ahr and ERR-target pyruvate dehydrogenase kinase 4 (Pdk4) (223) were mitigated by CH223191 and GSK5182, respectively. BPA-mediated decrease in GR- responsive gene early growth response 1 (Egr1) (224) was reversed by RU486. Finally, T0090709 has been previously shown to block BPA-mediated increase in Pomc expression in a hypothalamic cell line at the same concentration of BPA and T0070907 used in this study (203). These positive controls verified that the antagonists do have the potential to block BPA-mediated effects at the concentrations used in this study (Appendix B).

2.3.3 Anti-oxidant media and supplements

For DMEM versus Neurobasal A (NBA) experiments, cells were grown and treated as described above for 8 or 16 hours. For the NBA group, vehicle or 100 mM BPA was diluted 1:1000 in NBA medium, without phenol red (Gibco, cat. #12349015), supplemented with 1% CSFBS and 1% PS, instead of the phenol-red free DMEM (Hyclone, cat. #SH3028401).

N-acetylcysteine (NAC, cat. #A9165), pyridoxal hydrochloride (vitamin B6, cat. #P6155) and vitamin B12 (cat. #V6629) were purchased from MilliporeSigma and dissolved in sterile H2O to 500 mM, 20 mM and 500 M, respectively. 500 mM NAC was diluted 1:50 in phenol-red free DMEM to obtain a final concentration of 10 mM. As NAC changes the pH of media, the pH of media containing 10 mM NAC and media containing an equal volume of H2O was measured. pH of the 10 mM NAC media was matched to the H2O control by adding 5 M NaOH (approximately

3-4 L). An additional 3-4 L of H2O was added to the H2O containing media. Both H2O and 10 mM NAC containing media were then sterile filtered prior to pre-treating cells with H2O or NAC for 8 or 16 hours, followed by treatment with 100 M BPA or 0.05% EtOH for 8 hours or 16 hours. 20 mM vitamin B6 was diluted 1:1000 and 1:500 in media to final concentrations of 20

M and 40 M, respectively. 500 M vitamin B12 was diluted to 100 M (1:5) and 5 M (1:100) in H2O. Each of these was then diluted 1:1000 in media to achieve final concentrations of 500

41

nM, 100 nM and 5 nM vitamin B12. Vitamins B6 and B12 experiments were conducted as co- treatments with 100 M BPA or 0.05% EtOH exposure for 8 or 16 hours.

2.4 Quantitative RT-PCR

Total RNA was isolated with the PureLink RNA isolation kit (Thermofisher Scientific) according to the manufacturer’s instructions with genomic DNA removal using an on-column DNase step (PureLink DNase set, Thermofisher Scientific). RNA was quantified on the NanoDrop 2000 (Thermofisher Scientific) and 500 to 1000 ng of cDNA was synthesized using the Applied Biosystems High Capacity cDNA Reverse Transcription kit (Thermofisher Scientific). For the quantitative reverse transcriptase polymerase chain reaction (qRT-PCR), 12.5 ng of cDNA was amplified with gene specific primers (Table 1) and the Platinum SYBR Green qPCR SuperMix-UDF with ROX using the Applied Biosystems Prism 7900HT machine as described previously (225) with the following cycling conditions: 2 min at 50 °C, 2 min at 95 °C, 40 cycles of 15 s at 95 °C and 1 min at 60 °C, followed by melt curve analysis: 15 s at 95 °C, 15 s at 60 °C, 15 s at 95 °C. Data were analyzed with the standard curve method or with the ΔΔCT method and normalized to the reference gene, 60S ribosomal protein L7 (Rpl7).

Table 2-1: qPCR Primers Amplicon Gene Name Primer Sequence (5' --> 3') Size Agrp F: CGG AGG TGC TAG ATC CAC AGA 69 R: AGG ACT CGT GCA GCC TTA CAC Pre-Agrp F: TGA CCT CAG TCC ACT GCC A 139 R: AGG ACT CGT GCA GCC TTA CAC Ar F: CGC TGA AGG GAA ACA GAA GTA 120 R: AGA GTC ATC CCT GCT TCA TAA Ahr F: CCC TAC CAA TAC GCA CCA AA 104

R: GAG GGC ACT CAT AAG AGA ACT G Atf3 F: CTC CTG GGT CAC TGG TAT TTG 98 R: CCG ATG GCA GAG GTG TTT AT Atf4 F: GGA GCA AAA CAA GAC AGC AGC 179 R: TTG CCT TAC GGA CCT CTT CT Atf6 F: CAG ATG GTG ACA ACC AGA AAG A 112 R: CAT GGA GGT GGA GGC ATA TAA AG Bmal1 F: GGG AGG CCC ACA GTC AGA TT 78 R: GTA CCA AAG AAG CCA ATT CAT CAA

42

Catalase F: TCA CTG ACG AGA TGG CAC AC 168 R: ATC GAA CGG CAA TAG GGG TC Chop F: TAT GAG GAT CTG CAG GAG 109 R: CAG GGT CAA GAG TAG TGA AG Clock F: CAC CGA CAA AGA TCC CTA CTG AT 151 R: TGA GAC ATC GCT GGC TGT GT Cry1 F: AGA GCT CGG CTT TGA TAC AGA 120 R: CGT TCA AAG TTT GCC ACC CA Egr1 F: TAT ACT GGC CGC TTC TCC CT 114

R: AGA GGT CGG AGG ATT GGT CA Esr1 F: GAG TGC CAG GCT TTG GGG ACT T 102 R: CCA TGG AGC GCC AGA CGA GA Esr2 F: ATC TGT CCA GCC ACG AAT CAG TGT 114 R: TCT CCT GGA TCC ACA CTT GAC CAT Esrra F: GAG ACT GAG ACT GAA CCC CC 137

R: GAG CTG AGC ACT AGC TTC CC Esrr훾 F: ACT GTT GCA GTT GGA AAG GC 95

R: TGG AGG GTT CCG TCT TGA TGA Foxo1 F: CAA AGT ACA CAT ACG GCC 147 R: GAG AGT CAG AAG TCA ACA Gpx1 F: CGA CAT TGC CTG GAA CTT TG 118 R: GGA CAG CAG GGT TTC TAT GT Grp78 F: GCG ACA AGC AAC CAA AGA TG 117 R: TTC TTC TCT CCC TCT CTC TTA TCC Hmox1 F: TGA CAC CTG AGG TCA AGC AC 144 R: TCT GAC GAA GTG ACG CCA TC Hspb1 F: GGG CAC ACA TAA AAG CAC GC 101 R: CGG TCA TGT TCT TGG CTG GT Igf1 F: CCG TCC CTA TCG ACA AAC AAG A 141

R: TGT GGC ATT TTC TGC TCC GT Il6 F: GTG GCT AAG GAC CAA GAC CA 85 R: GGT TTG CCG AGT AGA CCT CA Il10 F: AGC ACT GCT ATG TTG CCT GCT CTT 95 R: TGA CTG GGA AGT GGG TGC AGT TAT Nfκb1 F: GGA TGA CAG AGG CGT GTA TTA G 114 R: CCT TCT CTC TGT CTG TGA GTT G Nos2 F: CCT GAA GGT GTG GTT GAG TT 124 R: CTT GGA AGA GGA GCA ACT ACT G Npy F: CAG AAA ACG CCC CCA GAA 77 R: AAA AGT CGG GAG AAC AAG TTT CAT T Npy #2 F: TAA CAA GCG AAT GGG GCT GT 71

43 for mHypoA-2/12 R: CAG CCA GAA TGC CCA AAC AC Nr3c1 F: TGT GAG TTC TCC TCC GTC CA 154 R: GTG CTG TCC TTC CAC TGC TC Nrf2 F: GGA CAT GGA GCA AGT TTG GC 164 R: CCA GCG AGG AGA TCG ATG AG Pdk4 F: AAA GAT GCT CTG CGA CCA GT 191

R: GGG TCA AGG AAG GAC GGT TT Per2 F: TCA TCA TTG GGA GGC ACA AA 135 R: GCA TCA GTA GCC GGT GGA TT Pomc F: CCC GCC CAA GGA CAA GCG TT 112 R: CTG GCC CTT CTT GTG CGC GT Ppar훾 F: GGT GAC TTT ATG GAG CCT AAG 110

R: CGG TCT CCA CTG AGA ATA ATG Rev-Erbα F: TGG AAG ACA GCA GCC GAG TG 114 R: CAT AGT GGA AGC CTC AGG CCA Rpl7 F: TCG CAG AGT TGA AGG TGA AG 114

R: GCC TGT ACT CCT TGT GAT AGT G Sod1 F: GGA ACC ATC CAC TTC GAG CA 87 R: CCC ATG CTG GCC TTC AGT TA Stat3 F: GCC ACG TTG GTG TTT CAT AAT C 97 R: TTC GAA GGT TGT GCT GAT AGA G Tfap2b F: TAC TCA CCT CAC TGG TAG AAG G 88 R: ACG GCT TTA GCA GGA AAC TC Tnf훼 F: CTC CTG GTA TGA GAT AGC 103 R: GTT GTA CCT TGT CTA CTC CC ChIP Primers Foxo1 site #1 F: GCA GCC ATT AAC ACT AAT GAA GC 143 R: TAC ACC TAT TTC CAT CCC CAC G Foxo1 site #2 F: GTG CCC TTG ACA AAG TTC CTG GAA 152 R: GCA GAA CCT AGG GAT GGG TCA TGC ATF:CRE site F: ATG GCA ACT GCC CCA AAC 132 R: AAG TAG TCT CCA CCC GGT T CEBP site F: TTT GAG GGA AAG GGG ATT GG 185 R: GTA AGA GCT AAT CCT AGG CGG T ATF3 on ATF3 F: CCA GTT CTC CTG GAA GCT A 110 promoter R: CGT TGC ATC ACC CCT TTT AA Npy ChIP F: TGT GCC TTC CTC CTT ATC AGA 103 R: GCC ACA AAC ACT GAG CTG TC

44

2.5 Western Blotting

2.5.1 Cell treatments

2.5.1.1 JNK and ERK signaling

For short term signaling experiments, cells were serum-starved in 2.5 mL of phenol-red free DMEM for 1 hour prior to treatment with 100 M BPA or 0.05% EtOH for 5, 15 or 30 minutes.

2.5.1.2 CHOP and ATF3 levels

Cells were grown to 80-90% confluency in DMEM supplemented with 2% FBS and 1% PS. On the day of treatment, cells were treated with 100 M BPA or 0.05% EtOH for 2, 4 or 8 hours in phenol-red free DMEM supplemented with 1% CS-FBS and 1% PS.

2.5.1.3 AMPK signaling

Cells were grown to 90% confluency in DMEM supplemented with 2% FBS and 1% PS. On the day of treatment, 2.5 mL of DMEM modified phenol-red free medium with 1% CSFBS and 1% PS was added to the cells 1 hour prior to treatment with 100 M BPA or 0.05% EtOH for 8 or 16 hours.

2.5.1.4 Validation of BMAL1-KO

Female and male mHypoA-Bmal1-KO and mHypoA-Bmal1–WT cell lines were grown to 90- 95% confluency and harvested.

2.5.2 Blotting

Protein was harvested using 1 x cell lysis buffer (Cell Signaling Technology Inc.) containing 1 mM phenylmethylsulfonyl fluoride, 1% protease inhibitor cocktail and 1% phosphatase inhibitor cocktail 2 (MilliporeSigma) or using the mirVANA PARIS kit (for siRNA validation, Thermofisher Scientific) and quantified using the BCA protein assay kit (Thermofisher Scientific). 15 - 25 g of total protein was separated on 10 or 12% SDS-polyacrylamide gels and transferred onto PVDF membranes (Bio-Rad). Membranes were blocked in 5% milk dissolved in tris-buffered saline with tween-20 (TBS-T) for 1 hour, prior to incubation with primary antibody overnight at 4°C. Primary antibodies CHOP (L63F7) (cat. # 2895), ATF3 (D2Y5W) (cat. #33593), ATF4 (D4B8) (cat. #11815), ATF6 (D4Z8V) (cat. #65880), -tubulin (cat #2144),

45 pJNK (Thr183/Tyr185, 81E11, cat. #4668), total JNK (cat. #9252), pERK1/2 (Thr202/Tyr204, cat. #9101), total ERK (137F5, cat. #4695), pAMPK (Thr172) (cat. #2532S), total AMPK (cat. #2535S) and BMAL1 (D2L7G) were purchased from Cell Signaling Technology Inc., and were diluted 1:1000 in 5% milk in TBS-T (for CHOP, JNK, ERK, AMPK, BMAL1) or in 5% bovine serum albumin (BSA) in TBS-T. Membranes were washed and incubated with secondary HRP-linked anti-mouse (for CHOP) or anti-rabbit antibody (1:7500 in 5% milk in TBS-T) (Cell Signaling Technology Inc.) for 1 hour, washed and imaged using the Signal Fire ECL Reagent (Cell Signaling Technology Inc.) on the Kodak Image Station 2000R. The Restore PLUS Western blot stripping buffer (Thermofisher Scientific) was used according to the manufacturer’s instructions before probing the same blot for subsequent proteins. Protein density was quantified using ImageJ.

2.6 siRNA knockdown

Cells were grown to 75-80% confluency in 100 mm tissue culture plates for transfection. siRNA duplexes (Table 2) and negative control (NC, cat. # 51-01-14-03) were purchased from Integrated DNA Technologies Inc. (IDT, Coralville, IA, USA). 25 nM of siRNA or negative control and 25 L/plate of Dharmafect 3 Transfection Reagent (Dharmacon, Cedarlane, Burlington, ON, Canada) were complexed for 20 minutes at room temperature in serum- and antibiotic-free DMEM (cat. #D5796, MilliporeSigma). The complexed reagents were diluted in 2% FBS containing antibiotic-free DMEM. Cells were incubated with 5 ml/plate of transfection media for 24 hours. The cells were then washed with PBS and treated with 100 M BPA or 0.05% EtOH for 8 hours (for all siRNA experiments in mHypoA-59 and mHypoE-41 cells) or for 16 hours (for Esr1 siRNA experiment in mHypoE-46 cells) in antibiotic-free phenol-red free DMEM containing 1% CS-FBS. Cells were lysed and RNA and protein were collected from the same plate using the mirVANA PARIS kit (Thermofisher Scientific) to measure changes in mRNA and protein levels as described above. For the Foxo1 and Esr1 siRNA experiments, siRNA knockdown was validated with qRT-PCR only as an appropriate antibody was not available.

46

Table 2-2: siRNA Duplex sequences

Gene Target siRNA Duplex sequence (5' --> 3') Target Region

Foxo1 mm.Ri.Foxo1.13.3 Exon 2

CUU CUG GAU AAU CUC AAC CUU CUC T

AGA GAA GGU UGA GAU UAU CCA GAA GGU

Chop mm.Ri.Ddit3.13.3 Exon 3/4

AAC AGA GGU CAC ACG CAC AUC CCA A

UUG GGA UGU GCG UGU GAC CUC UGU UGG

Atf3 mm.Ri.Atf3.13.3 Exon 2

CUG GAG UCA GUU ACC GUC AAC AAC A

UGU UGU UGA CGG UAA CUG ACU CCA GCG

Atf4 mm.Ri.Atf4.13.1 Exon 3/4

AAG ACU GAG AAA UUG GAU AAG AAG C

GCU UCU UAU CCA AUU UCU CAG UCU UCA

Atf6 mm.Ri.Atf6.13.2 Exon 15

GAC CAA AAA UGU CAA UUG UAU UAC C

GGU AAU ACA AUU GAC AUU UUU GGU CUU

Esr1 mm.Ri.Esr1.13.1 Exon 5

AAU GAU GGG CUU AUU GAC CAA CCTA

UAG GUU GGU CAA UAA GCC CAU CAU UGA

47

2.7 Esr1 overexpression mHypoE-46 cells were grown to 75% confluence in 60 mm tissue culture plates for transfection. An estrogen receptor alpha (ER) overexpression plasmid (Esr1_OMu23198C_pcDNA3.1(+)) was designed by and purchased from GenScript USA, Inc. (Piscataway, NJ, USA). Briefly, the Esr1 open reading frame (OMu23198/NM_007956.5) was cloned into the pcDNA.3.1(+) vector using HindIII and BamHI restriction sites and Kozak sequences were added to ensure protein translation. 50 ng of the ER overexpression plasmid or the control plasmid (pcDNA.3.1(+)) and 6 L/plate of TurboFect Transfection Reagent (Thermofisher Scientific, cat. #R0531) were complexed for 20 minutes at room temperature in serum- and antibiotic-free DMEM (MilliporeSigma, cat. #D55796). Growth media was replaced with 3 mL/plate of antibiotic-free DMEM containing 2% FBS and 200 L of the transfection reagent/plasmid complex was added drop-wise to each plate. After 6 hours, transfection media was removed, cells were washed with PBS and treated with 100 M BPA or 0.05% EtOH for 16 hours in antibiotic-free, phenol-red free DMEM with 1% CSFBS. RNA was isolated as described below for qRT-PCR.

2.8 In silico promoter analysis and Chromatin immunoprecipitation (ChIP)

2.8.1 ChIP for ATF3

The mouse Agrp regulatory region was obtained, and the transcriptional start site was inferred from previously published papers on the Agrp promoter (133). ATF:CRE (TGACGTCA) and CEBP (TGT/ATGCAAT) sites were manually identified (226). mHypoA-59 cells were grown to 80-85% confluency in 100 mm tissue culture plates for treatment. Cells were treated with 100 M BPA or 0.05% EtOH for 1, 4 or 8 hours. An additional plate of cells was treated with 0.05% EtOH (vehicle) per experimental replicate for immunoprecipitation with the negative control (normal rabbit IgG) antibody. Cells were harvested in PBS containing 1 x protease inhibitor cocktail and ChIP was performed using the SimpleChIP Enzymatic Chromatin IP kit with magnetic beads (Cell Signaling Technology Inc.) according to the manufacturer’s instructions. 2 μg of ATF3 antibody (D2Y5W, cat. #33593, Cell Signaling Technology Inc.) or normal rabbit IgG antibody (provided with the kit) were used for

48 the immunoprecipitation. DNA was purified using columns provided with the SimpleChIP kit or the PureLink PCR Purification Kit (Thermofisher Scientific). Binding of ATF3 to the Agrp promotor region was analyzed using qRT-PCR with cycling conditions as described previously (225). Each sample was run in triplicate using 3.5 L of purified DNA in 10 L reactions with primers specific for the Agrp promotor or the Atf3 promoter (Table 1) and the Platinum SYBR Green qPCR SuperMix-UDF with ROX. Mean cycle of threshold (CT) of each immunoprecipitation sample (IP sample) and its respective 2% input sample were used to calculate the percent of DNA pulled down by the antibody relative to input: % of input = 2% x 2(CT 2% input sample – CT IP sample). Relative binding was calculated by taking the average of the percent inputs of vehicle and treated groups per experimental replicate and dividing each value by the respective average.

2.8.2 ChIP for BMAL1

Sequences of the 2500 base pairs upstream of the transcriptional start site of Npy and Agrp were obtained. The start site of each gene was determined using NCBI GenBank (Genome assembly GRCm38.p4, Annotation release 106): Npy start Chromosome 6, 49822710 and Agrp start Chromosome 8, 105566695 (reverse strand). BMAL1:CLOCK binding sites (5’CACGTG 3’, 5’CACGNG3’, 5’CACGTT3’, 5’CATG(T/C)G3’ or 5’CANNTG3’) were manually identified. Due to an updated genome annotation, relative positions of binding sites slightly differ for those that were published previously for Npy and Agrp (167). For instance, site 5’CATGTG3’ at -1226 in the Npy promotor in Fick et al. 2010 (167, 203) is listed here as -1207. mHypoA-Bmal1-WT/F cells were grown to 80-85% confluence in 10 cm tissue culture dishes and treated with vehicle (0.05% EtOH) or 100 μM BPA for 4 hours as described above. Additional dishes of cells were treated with vehicle for immunoprecipitation with positive (Histone H3) and negative control (normal rabbit IgG) antibodies. ChIP was performed using the SimpleChIP Enzymatic Chromatin IP kit with magnetic beads (Cell Signaling Technology, Inc.) following the manufacturer’s instructions. 2 g of BMAL1 antibody (D2L7G, Cell Signaling Technologies, Inc.) was used for the immunoprecipitation. This was the same antibody used to verify the presence or absence of BMAL1 protein in the cell lines. Purified DNA was analyzed via quantitative PCR on an Applied Biosystem Prism 7900HT machine with cycling conditions as described above. Each sample was assayed in duplicate using 2 μL of DNA in a 20 μL

49 reaction with primers specific for the Npy promotor (-1267 to -1165 relative to the transcriptional start site) (Table 1) or the Rpl30 promotor (for positive control H3 provided with the kit). The primer for the Npy promotor was designed previously and the region validated to bind BMAL1 (167). Mean cycle of threshold of each immunoprecipitation sample (IP sample) and its respective 2% input sample were used to calculate the amount of DNA pulled down using the following equation: percent input = 2% x 2(C(T) 2% input sample – C(T) IP sample). Relative binding was calculated by taking the average of the percent inputs of both vehicle and treated groups per experimental replicate and dividing each value by the respective average.

2.9 Statistical Analysis

Data were analyzed for statistical significance using GraphPad Prism 6.0 (GraphPad Prism Software Inc., San Diego, CA, USA). A student T-test or a one-way or two-way ANOVA, followed by the Bonferroni or Tukey multiple comparison test, was performed as appropriate and is indicated in the figure legends. Differences were considered statistically significant when P < 0.05. Data are presented as mean ± SEM, with statistical significance denoted as *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 when comparing vehicle versus BPA treated groups, as #P < 0.05, ##P < 0.01, ###P < 0.001, ####P < 0.0001 when comparing the effect of inhibitor pre-treamtent and co-treatment, siRNA knockdown or different medias within vehicle or BPA treated groups, and +P < 0.05, ++P < 0.01, +++P < 0.001, ++++P < 0.0001 when comparing the interaction between two factors. The inhibitor, supplement or siRNA was deemed to have an effect if the interaction was significant. ‘n’ values refer to independent experimental replicates.

Chapter 3 Bisphenol A induces Agrp gene expression in hypothalamic neurons through a mechanism involving ATF3

Manuscript published in Neuroendocrinology

Citation: Loganathan N, McIlwraith EK, Belsham DD. Bisphenol A induces Agrp gene expression in hypothalamic neurons through a mechanism involving ATF3. Neuroendocrinology 2020, doi: 10.1159/000509592

Contributions: • NL completed the experiments and wrote the manuscript • EKM assisted with or completed experiments in Fig 3-1C, 3-1E • DDB provided scientific input, direction and funding • All authors edited the manuscript

50 51

3 Bisphenol A induces Agrp gene expression in hypothalamic neurons through a mechanism involving ATF3 3.1 Abstract

Bisphenol A is a ubiquitous endocrine disrupting chemical and obesogen. Although limited evidence exists of the effects of BPA on hypothalamic AgRP levels, the mechanisms underlying these effects remain unknown. Given that AgRP is a potent orexigenic neuropeptide, determining the mechanism by which BPA increases AgRP is critical to preventing the progression to metabolic disease. Using qRT-PCR, we investigated the response of Agrp-expressing mouse hypothalamic cell lines to BPA treatment. The percentage of total BPA entering hypothalamic cells in culture was quantified using an enzyme-linked immunosorbent assay. In order to identify the mechanism underlying BPA-mediated changes in Agrp, siRNA knockdown of transcription factors, FOXO1, CHOP, ATF3, ATF4 and ATF6, and small molecule inhibitors of endoplasmic reticulum stress, JNK or MEK/ERK were used. BPA increased mRNA levels of Agrp in six hypothalamic cell lines (mHypoA-59, mHypoE-41, mHypoA-2/12, mHypoE-46, mHypoE-44, mHypoE-42). Interestingly, only 18% of the total BPA in the culture media entered the cells after 24 hours, suggesting exposure concentration is much lower than treatment concentration. BPA increased pre-Agrp mRNA levels, indicating increased Agrp transcription. Knockdown of the transcription factor ATF3 prevented the BPA-mediated increase in Agrp, pre-Agrp and in part, Npy, mRNA levels. However, chemical chaperone, sodium phenylbutryate, JNK inhibitor, SP600125, or MEK/ERK inhibitor, PD0352901, did not block BPA-induced Agrp upregulation. Overall, these results indicate hypothalamic Agrp is susceptible to dysregulation by BPA and implicate ATF3 as a common mediator of the orexigenic effects of BPA in hypothalamic neurons.

3.2 Introduction

Obesity is a pressing health concern as it leads to complications, including cardiovascular disease, diabetes and some cancers. Although calorie-dense diets and a sedentary life style are major reasons for the recent rise in obesity, the contribution of environmental endocrine disrupting chemicals, termed “obesogens”, has recently gained attention (15, 227). BPA, a chemical widely used in polycarbonate plastics and receipt paper, has been correlated to

52 increased BMI and diabetes (45, 47, 228, 229), increases body weight in rodent models (60), and has recently been classified as a metabolism disrupting chemical (MDC) (227). BPA is detected in the urine of 9 in 10 people (228), and accumulates in tissues, including the brain (37).

Although the obesogenic effects of BPA have been primarily described as targeting fat cells, leading to adipogenesis, lipogenesis and ultimately weight gain (44, 60, 65), its effects on the hypothalamic control of energy homeostasis have been largely neglected. The hypothalamus houses orexigenic NPY/AgRP neurons and anorexigenic POMC neurons that integrate peripheral indicators of energy status to maintain energy homeostasis (16). Exogenous chemicals that affect these neurons can override this intricate control of energy balance, leading to dysregulation. In fact, maternal exposure to BPA increased protein levels of AgRP and decreased POMC in neural progenitor cells of newborn mice (184). McKay et al. also described that in male offspring, perinatal exposure to BPA led to decreased POMC fiber innervation and increased Npy and Agrp mRNA levels when combined with a high fat diet (57). The molecular mechanisms underlying these changes remained unknown, and these paradigms do not address the effects of direct BPA exposure on neurons in the mature hypothalamus. The chronic activation of orexigenic NPY and AgRP will ultimately lead to obesity as demonstrated with ICV administration of the neuropeptides (16). Notably, a single dose of AgRP was shown to increase food intake in a rodent model for seven days, highlighting its potency (86).

We have described an increase in Npy, Agrp (Chapters 3, 4) and Pomc mRNA expression with BPA exposure in mouse hypothalamic neurons in vitro, suggesting an overall dysregulation of the hypothalamic melanocortin system beginning at the transcriptional level (203, 225). Mechanistically, as described in chapter 5, the increase in Npy mRNA levels was mediated by the circadian transcription factor BMAL1 (225), while the transient upregulation of Pomc at 4 hours was mediated by neuroinflammation and the transcription factor PPARγ (203). However, blocking these pathways did not block the BPA-mediated upregulation in Agrp mRNA, suggesting another distinct pathway is involved in mediating the changes in Agrp. Given that AgRP is a potent, orexigenic neuropeptide, determining the mechanism by which BPA increases its transcription is critical to blocking the orexigenic effects of BPA.

As an orexigenic peptide, Agrp is suppressed by signals of high energy, including insulin (131), leptin (133), high glucose (139) and amino acids (142). These signals ultimately regulate Agrp

53 through a set of transcription factors. An understanding of the transcriptional regulation of mouse Agrp is limited, in part due to its large regulatory region that lies up to 43 kb upstream of its transcriptional start site (145). Nevertheless, FOXO1 is a well-known activator of Agrp transcription, while STAT3 directly represses Agrp transcription (133). We therefore hypothesized that BPA increases Agrp mRNA levels by activating FOXO1 or repressing STAT3. Many other transcription factors are likely involved in the promoter regulation of Agrp. Besides circadian dysregulation and neuroinflammation, BPA has been described to induce EndR stress (207). EndR stress occurs as a result of elevated protein misfolding, and also occurs in states of nutrient deprivation (230) and obesity (231), both of which are accompanied by changes in Agrp. A number of transcription factors are involved in EndR stress, such as CHOP, ATF3, ATF4 and ATF6 (230), that may be involved in Agrp regulation by direct binding to the Agrp regulatory region or through interaction with other Agrp-associated transcription factors. For example, ATF3 has been previously shown to bind dimerize with FOXO1 on the Agrp promoter, leading to increased transcription (141). Therefore, we also hypothesized that BPA increases Agrp mRNA levels by altering EndR stress-regulated transcription factors.

Herein, we describe that BPA does indeed lead to transcriptional upregulation of Agrp mRNA in both adult and embryonic-derived mouse hypothalamic cell lines, with concurrent upregulation in the expression of Foxo1, Stat3, Chop, Atf3, Atf4 and Atf6. We further identify that the increase in Agrp is dependent on the transcription factor ATF3, yet independent of FOXO1 activation and EndR stress. We also describe that ATF3 is, in part, involved in BPA-mediated Npy upregulation. Overall, these results illustrate a FOXO1-independent action of ATF3 on Agrp and Npy and highlight a common mechanism to mitigate the orexigenic neuropeptide increasing effects of BPA in the hypothalamus.

3.3 Results

3.3.1 BPA increases Agrp mRNA levels in several hypothalamic cell lines To characterize the effects of BPA and establish the optimal treatment dose, the adult-derived mHypoA-59 cells were treated with 10 or 100 M BPA for 4 hours. 100 M BPA increased Agrp expression in mHypoA-59 cells (Fig 3-1A). 10 M BPA did not alter Agrp expression in the embryonic-derived mHypoE-41 cells; however, 25, 50 and 100 M BPA increased Agrp mRNA in this cell line (Fig 3-1B). These results suggested that BPA does act to induce

54 orexigenic neuropeptide expression in both adult- and embryonic- derived hypothalamic neurons, albeit at a higher concentration than what is reported in human serum (nM levels) (36, 232).

For this reason, we aimed to determine the amount of BPA that was able to enter the neurons in culture. BPA may not readily dissolve in aqueous culture media and enter cells due to its lipophilic nature (233). To determine the percentage of BPA in the media that enters hypothalamic neurons, an ELISA was performed to determine the concentration of BPA within the cells and in the culture media 4 and 24 hours after treatment. At 4 hours, 2% and 10% of the total measured BPA was found within the cells when treated with 10 and 100 M BPA, respectively. At 24 hours, 14% with the 10 M treatment and 18% with the 100 M treatment of the total BPA content was found within the cells (Fig 3-1C). Thus, only a fraction of the BPA that is dissolved into the culture media actually enters the hypothalamic cell, indicating a lower exposure concentration than the apparent treatment concentration. Furthermore, with lower treatment concentration, there is lower percentage of entry into the cells. Accordingly, when mHypoA-59 cells were treated with 10 and 100 M BPA in Fig 3-1A, only the higher apparent concentration of 100 M was able to increase Agrp expression. We therefore chose to use 100 M BPA in our subsequent experiments to ensure adequate and consistent entry of BPA into hypothalamic cells in culture.

Treatment with 100 M BPA for 8 hours in hypothalamic primary culture derived from 8-week old male mice (Fig 3-1D), and in six hypothalamic cell lines from 2 to 24 hours (Fig 3-1E) revealed consistent upregulation of Agrp across cell models, again illustrating the orexigenic potential of BPA across an array of AgRP neurons. In embryonic-derived cell lines (mHypoE-), the upregulation occurred from as early as 2 hours until 8 or 16 h, whereas in adult cell lines (mHypoA-), Agrp was significantly upregulated at 8 hours alone (Fig 3-1E).

55

10 µM BPA A mHypoA-59 B mHypoE-41 C mHypoA-59 100 µM BPA D Primary Culture

2 2 t 25 ** 1.5

** n

e

t

A

A

A n

N * 20 * N **

N * o

* R

c

R

R

m 1.0

m

A

m

15

7

P

7

l

7

l l

1 B

p p

1

p

l

R

R

a

/ R

10 /

t

/

p

p o

p 0.5

r

r

t

r

g

g

f

g

A A

o 5

A

% 0 0 0 0.0 Veh 10 100 Veh 10 25 50 100 4 h 24 h Veh100 µM BPA µM BPA µM BPA

E 100 µM BPA

2 **** ** ** ** *** * *** * ** ***

A ** *

N * ***

R

m

7

l 1

p

R

/

p

r

g A

0 hours 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 mHypoA-59 mHypoE-41 mHypoA-2/12 mHypoE-46 mHypoE-44 mHypoE-42

Cell Line

Figure 3-1 BPA increases Agrp mRNA expression in hypothalamic cell lines. (A) mHypoA-59 cells treated with 10 or 100 µM BPA or vehicle (0.05% EtOH) for 4 hours (n=3). (B) mHypoE-41 cells treated with 10, 25, 50 or 100 µM BPA or vehicle for 4 hours (n=4). (C) mHypoA-59 cells treated with 10 or 100 µM BPA or vehicle for 4 or 24 hours, followed by collection of both cells and media to determine BPA content (n=3-4). (D) Primary culture derived from the hypothalamus of male CD-1 mice treated with 100 µM BPA or vehicle for 8 hours (n=4). (E) Six Agrp-expressing cell lines treated with 100 µM BPA or vehicle for 2,4, 8, 16 or 24 hours (n=3-6). Time-matched vehicles are represented by the dotted line at y=1 and stars indicate a significant difference compared with the time-matched vehicle. Agrp gene expression was analyzed using qRT-PCR. Data are expressed as mean +/- SEM, and statistical significance was determined using a student T-test (D); or a One-way ANOVA (A, B) or a Two-way ANOVA (C, E), followed by the Bonferroni post-hoc test (A, B, C, E) or the Tukey multiple comparison test (C); *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001

56

3.3.2 BPA increases transcription of Agrp mRNA and associated transcription factors

The mechanism underlying BPA-induced Agrp expression may be a result of transcription factor binding and transcriptional activation at the promotor level, or a result of increased stability of the transcript if there is no de novo transcription occurring. To test this, the transcriptional inhibitor, actinomycin D (ActD), was used prior to treatment with BPA. ActD decreased basal and BPA-induced levels of Agrp at 8 hours in the mHypoA-59 and mHypoE-41 cell lines, indicative of the inhibition of de novo Agrp transcription and subsequent mRNA degradation. Furthermore, in the presence of ActD, BPA did not increase Agrp mRNA levels compared to vehicle, suggesting that transcription is required for the BPA-induced upregulation of Agrp (Fig 3-2A, B). To determine if Agrp itself is regulated at the transcriptional level, changes in a region containing an intronic portion of Agrp (pre-mRNA) were measured (Fig 3-2C). The expression of pre-Agrp mRNA was upregulated by BPA at 2 hours in the mHypoA-59 cells (Fig 3-2D) and at 2 and 8 hours in the mHypoE-41 cell line (Fig 3-2E), suggesting de novo transcription of Agrp.

Given that BPA increases transcription of Agrp, we hypothesized that the mechanism of BPA action depends on one or more transcription factors that regulate Agrp at the promoter level. We took two approaches to answer this question. First, we investigated transcription factors that are known to regulate Agrp, Foxo1 and Stat3 (133). Second, we had previously found that BPA- induced pathways, including steroid receptor activation (Appendix A) circadian dysregulation (Chapter 5.3.4) and oxidative stress (Appendix C), did not mediate its effects on Agrp. Endoplasmic reticulum stress (EndR stress) is often activated by BPA (207) and is tightly linked to hypothalamic dysregulation seen in obesity (231). Furthermore, EndR stress activates a myriad of transcription factors that may activate Agrp transcription directly or interact with Agrp-activating transcription factors (230). As such, we first determined whether these transcription factors are altered with BPA treatment. In both the mHypoA-59 and mHypoE-41 cells, Foxo1 and Stat3 as well as stress-responsive transcription factors, Chop, Atf3, Atf4 and Atf6 were increased with 100 M BPA treatment from 2 to 8 hours (Fig 3-2F). Atf3 and Chop had the greatest fold-change of approximately 10-fold (both cell lines) and 8 or 5-fold (mHypoA-59 and mHypoE-41, respectively) at 8 hours (Fig 3-2F). Protein levels of both CHOP and ATF3 were also increased, indicating their potential to bind to and alter transcription at the promoter level

57

(Fig 3-2G, H). As Stat3 levels were only modestly increased, rather than decreased as would have been expected from previous studies (133), we did not follow up on the potential role of this transcription factor in BPA-mediated Agrp upregulation.

C Agrp gene Pre-Agrp For Agrp For Pre-Agrp and Agrp Rev 5' 3' A mHypoA-59 B mHypoE-41 Exon 4 Exon 5 ++ intron ++ D mHypoA-59 E mHypoE-41

2 2 *** 2 2

A A

A

A

N N N

** * N

R R R

R *** ***

m

m

m

m

7 7

7

7

l

l l l p

1 1 p

p

p

R

R R

R 1 1

/

/

/

/

p

p

p

p

r

r r

r ### #### g

## g g

g ####

A

A

-

-

A

A

e

e r

0 0 r

P P DMSO ActD DMSO ActD 0 0 hours 2 4 8 2 4 8 F 12 mHypoA-59 **** mHypoE-41 ****

A **** N

R 8 **** ****

m

7 l ****

p **** ****

R /

e 4 **** **** **** *** **** n ****

e *** **** *** ** ** *** *** **** G ** ** ** ** ******* 0 hours 2 4 8 2 4 8 2 4 8 2 4 8 2 4 8 8 2 4 8 2 4 8 2 4 8 2 4 8 2 4 8 8 Foxo1 Stat3 Chop Atf3 Atf4 Atf6 Foxo1 Stat3 Chop Atf3 Atf4 Atf6 G mHypoA-59 H mHypoA-59 4 *

3 n

i Legend

l n

u 3

i l

b mHypoA-59

u

u

) t b 2

- Veh

D

u t

α 2

- /

S 100 µM BPA

α

-

P

/

/ 3

+ mHypoE-41

O (

F 1

H 1

T Veh C A 100 µM BPA 0 0 2 4 8 2 4 8 Hours Hours CHOP ATF3

α-tubulin α-tubulin

BPA - + - + - + BPA - + - + - +

58

Figure 3-2 BPA increases pre-Agrp mRNA levels and associated transcription factors. (A) mHypoA-59 or (B) mHypoE-41 cells pre-treated with transcriptional inhibitor 10 µg/ml actinomycin D (ActD) or 0.1% DMSO for 1 hour, followed by treatment with 100 µM BPA or vehicle (0.05% EtOH) for 8 hours (n=4). (C) Schematic of the locations of primers used to amplify mature Agrp and pre-Agrp using qRT-PCR. (D-F) mHypoA-59 or mHypoE-41 cells treated with 100 µM BPA or vehicle for 2, 4 or 8 hours (same samples as Fig 1E) (n=3-5). Time-matched vehicles are represented by the dotted line at y=1 and stars indicate a significant difference compared with the time-matched vehicle. Gene expression was analyzed using qRT-PCR. (G-H) mHypoA-59 cells treated with 100 µM BPA or vehicle for 2, 4 or 8 hours for protein quantification (n=3-4 for CHOP, n=2 for ATF3). Protein levels were analyzed using western blotting and representative blots from one experimental replicate are shown. No bands for CHOP or ATF3 appeared in vehicle treated samples, thus only BPA treated samples were quantified by densitometry. Data are expressed as mean +/- SEM (except H), and statistical significance was determined using a Two-way ANOVA, followed by the Tukey multiple comparison test (A, B) or the Bonferroni post-hoc test (D, E, F), or a One-way ANOVA, followed by the Bonferroni post- hoc test (G). Veh vs BPA: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; DMSO vs. ActD: ##P<0.01, ###P<0.001, ####P<0.0001; interaction: ++P<0.01

3.3.3 Knockdown of ATF3 blocks the BPA-induced increase in Agrp and pre-Agrp mRNA

To determine whether the BPA-mediated increase in Agrp was dependent on any of the above- mentioned transcription factors, siRNA-mediated knockdown was performed (Fig 3-3A), and transcription and translation of each transcription factor was significantly decreased with siRNA exposure. For the Foxo1 siRNA, knockdown was confirmed only at the transcript level as an appropriate antibody was not available. Knockdown of ATF3 in the mHypoA-59 cells increased basal expression of Agrp, but blocked an increase in Agrp with BPA (Fig 3-3B), suggesting ATF3 is necessary for the BPA-mediated orexigenic effect. Interestingly, knockdown of FOXO1, CHOP, ATF4 and ATF6 did not prevent the upregulation of Agrp by BPA (Fig 3-3B). ATF3 knockdown also prevented BPA-mediated Agrp induction in the embryonic-derived, mHypoE-41 cell line (Fig 3-3C, D). Similar results were seen with pre-Agrp mRNA (Fig 3-3E), suggesting ATF3 is required for BPA-induced de novo transcription of Agrp.

59

A mHypoA-59 siRNA validation Foxo1 Chop Atf3 Atf4 Atf6 ++++ 7 ****

+++ A

N 6

R 4 ++++

m ****

++ 7

l 3

p R

/ ****

e 2 ****

n #### ####

e ####

G 1 ## *** #### #### #### * #### #### 0 NC Foxo1 NC Chop NC Atf3 NC Atf4 NC Atf6 siRNA siRNA siRNA siRNA siRNA ATF4 CHOP ATF3 ATF6

α-tubulin B mHypoA-59 Agrp ++

2

A p=0.06 ** # ** ** N ** * ** **

R *

m

7 l

p 1

R

/

p

r

g A 0 NC Foxo1 NC Chop NC Atf3 NC Atf4 NC Atf6 siRNA siRNA siRNA siRNA siRNA

C mHypoE-41 D mHypoE-41 E mHypoA-59 mHypoE-41 siRNA validation Agrp Pre-Agrp ++++ ++

+ + A

****

N

8 A A **** R

N 2 2 # N

m **

R

R 7

6 m

l

m

p

7

l

7

R

l p

#### /

p p

4 R /

1 r 1 R

/ ***

p

g

r

3

A

f

g

t

- A

2 e

A

r P 0 0 0 NC Atf3 NC Atf3 NC Atf3 NC Atf3 siRNA siRNA siRNA siRNA ATF3 Legend mHypoA-59 mHypoE-41 Veh Veh α-tubulin 100 µM BPA 100 µM BPA

60

Figure 3-3 ATF3 knockdown prevents BPA-mediated increase in Agrp and pre-Agrp mRNA. (A, B, E) mHypoA-59 cells transfected with siRNA duplexes targeting Foxo1, Chop, Atf3, Atf4 or Atf6 or negative control (NC) for 24 hours, followed by treatment with 100 µM BPA or vehicle (0.05% EtOH) for 8 hours (n=3-5). (C, D, E) mHypoE-41 cells transfected with a siRNA duplex targeting Atf3 or NC for 24 hours, followed by treatment with 100 µM BPA or vehicle for 8 hours (n=6). siRNA knockdown of the various transcription factors (A) or ATF3 (C) was confirmed using qRT-PCR and western blotting prior to analyzing changes in Agrp (B,D) or pre-Agrp (E) mRNA levels using qRT-PCR. Representative blots from one experimental replicate of the siRNA knockdown validation are shown below the qRT-PCR validation (A,C). Data are expressed as mean +/- SEM, and statistical significance was determined using a Two-way ANOVA, followed by the Tukey multiple comparison test (A-E); Veh vs. BPA: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; NC vs. siRNA: #P<0.05, ##P<0.01, ####P<0.0001; interaction: +P<0.05, ++P<0.01, +++P<0.001, ++++P<0.0001

3.3.4 BPA does not increase ATF3 binding to Agrp regulatory elements

ATF3 is a transcription factor that can either bind ATF:CRE elements, or dimerize with other transcription factors, including FOXO1, CEBP, p53 and CREB among several others (226). The ATF3:FOXO1 dimer has been previously validated to bind to and increase transcription of Agrp in response to low glucose conditions (141). Alternatively, ATF:CRE, AP1 and CEBP sites were found to be the most common ATF3 binding sites in the entire genome (226). Several AP1 binding sites are present in the Agrp regulatory region. We identified a potential ATF:CRE (TGACGTCA) and a potential CEBP (TGTTGCAAT) site in the distal regulatory region of Agrp (Fig 3-4A).

ChIP experiments revealed that increased ATF3 binding to the previously validated FOXO1 binding sites on Agrp did not occur with 100 M BPA treatment for 1 or 4 hours (Fig 3-4B, C). Relative binding of ATF3 was greater than IgG at the ATF:CRE site at 1 hour, suggesting a basal degree of ATF3 binding to this region. However, BPA did not increase ATF3 binding at the ATF:CRE site or the CEBP site in the Agrp distal regulatory region (Fig 3-4B). Thus, an alternate binding site within the large 43 kb Agrp promoter is likely involved. ATF3 is known to bind to its own promoter, which was thus used as a positive control for the antibody with previously validated primers (234). With 4 or 8 hours of BPA treatment, ATF3 binding to the ATF3 promoter was increased (Fig 3-4C).

61

A Agrp regulatory region

ATF:CRE site CEBP site FOXO1 site 1 FOXO1 site 2 5' 3' TGACGTCA TGTTGCAAT TGTTT AAATA AAACA Agrp

-38,306 -31,306 -356 -299 -106 +1 CATCCC

B 1 hour Legend C 4 hours/8 hours Veh - ATF3 mHypoA-59 100 µM BPA - ATF3 mHypoA-59 *

2 IgG 4

g

g

n

n

i

i d d 3

n *

n

i

i

b

b

e

1 e 2

v

v

i

i

t

t

a

a

l

l e

e 1

R R 0 0 FOXO1 FOXO1 ATF:CRE CEBP FOXO1 FOXO1 ATF:CRE ATF3 site on site 1 site 2 site site site 1 site 2 site ATF3 promoter

Figure 3-4 BPA does not increase ATF3 binding to specific Agrp regulatory regions. (A) Schematic of transcription factor binding sites in the Agrp regulatory region. Circles and sequences represent potential ATF3 binding sites. Boxes represent the sites encompassed by each primer pair used for chromatin immunoprecipitation (in 4B, C). (B, C) mHypoA-59 cells treated with 100 µM BPA or vehicle (0.05% EtOH) for 1 hour (n=3-5) (B) or 4 hours (n=3) (C), followed by chromatin immunoprecipitation with an ATF3 antibody or IgG. Samples were treated for 4 or 8 hours to determine binding of ATF3 to the ATF3 promoter. Level of ATF3 or IgG binding to each region was assayed using qRT-PCR. Data are expressed as mean +/- SEM, and statistical significance at each site was determined using a One-way ANOVA, followed by the Bonferroni post-hoc test (B, C); *P<0.05

3.3.5 BPA induces EndR stress, JNK, ERK and AMPK, but does not require EndR stress, JNK or ERK to increase Agrp mRNA

The factors upstream of ATF3, leading to its induction, and ultimately the increase in Agrp, remain unknown. Treatment with ActD suggests that transcription of a gene(s) is required for the increase in Atf3 (Fig 3-5A). MAP kinases and EndR stress are well-characterized regulators of ATF3, with ATF4 being the transcription factor directly upstream of ATF3 in the EndR stress pathway (235, 236). Interestingly, in the mHypoA-59 cells, although ATF4 knockdown decreased basal levels of ATF3, it did not prevent the BPA-mediated increase in ATF3 (Fig 3- 5B), suggesting that akin to Agrp (Fig 3B), BPA acts independently of ATF4 to upregulate Atf3.

62

We next investigated whether common upstream regulators of ATF3 may mediate the BPA- induced changes in Agrp. As BPA induces EndR stress (Fig 3-2F), we used the generalized chemical chaperone, sodium phenylbutyrate (PBA) to mitigate EndR stress. 78-kDa glucose- regulated protein (GRP78), which is increased with EndR stress (237), was used as a positive control to establish the EndR stress-inhibitory activity of PBA (Appendix B). However, PBA did not significantly abolish the BPA-mediated upregulation of Agrp (Fig 3-5C), suggesting this effect is independent of EndR stress. MAP kinases JNK and ERK were also induced with BPA (Fig 3-5D, 5E, 5G), but inhibition of each using SP600125 (JNK) or PD0352901 (ERK) did not block the increase in Agrp with BPA (Fig 3-5F, H). Finally, the energy sensor, AMPK, can upregulate ATF3 (140). Remarkably, BPA increased AMPK phosphorylation at 8 hours in the mHypoA-59 cells (Fig 3-5I), and AMPK activation using 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR) increased Agrp expression. Lastly, treatment with 6 M Compound C, an AMPK inhibitor, prevented the BPA-mediated increase in Atf3 and Agrp in both mHypoA-59 and mHypoE-41 cells (Fig 3-5K,L), pointing to a molecular link between AMPK, ATF3 and Agrp.

63

A mHypoA-59 mHypoE-41 B mHypoA-59 C mHypoA-59 mHypoE-41 ++++ ns 3 ++++

**** ns

14 A

A N

12 A **** ***

N R **** N 2 ## *

R 2 R

10 m *

m

m **** 7

** l

8 **

7

p

7 l

2 l

p

R

p

/ R

R 1 p

/ 1

/

r

3

3

f

g f

t 1 t

### A A

#### A ## #### ## 0 0 0 DMSO ActD DMSO ActD NC Atf4 H2O 5 mM 10 mM H2O 5 mM 10 mM siRNA PBA PBA PBA PBA D mHypoA-59 E mHypoE-41

F mHypoA-59 mHypoE-41

2 2

K ** K

N **** *** N

A **

J J

***

N

l

l 2

a R

a t

t #

o m

o *** t

1 1

t

/ *

/

7

l

K

K

p

N N

R 1

J

/

J

p

p

p r

0 0 g

15 min 30 min 15 min 30 min A 0 pJNK pJNK 0.1% 50 µM 0.1% 50 µM DMSO SP DMSO SP total total JNK JNK G mHypoA-59 H mHypoA-59 mHypoE-41 I mHypoA-59 J mHypoA-59

*** K

1.5 A 1.5 2

A P

K *** N ****

2 N

R R

** M R

E **

A

m

*

l

l m

1.0 * 1.0

a

7

a

l

t

t 7

l

p

o o

t 1

p

t

/ R

1 /

/

R

K

/ K

0.5 p 0.5

p

R

r

P

r

g

E

g

M

p

A

A A

0.0 0 p 0.0 0 Veh 100 uM 0.1% 10 µM 0.1% 10 µM Veh 100 µM Veh 500 µM BPA DMSO PD DMSO PD BPA AICAR pERK pAMPK total total ERK AMPK

K ++++ L Legend

++++ 2 ++++ ++++ A

A 10 **** mHypoA-59

N

N R R **** **** Veh

**** m

m

10 µM BPA

7

7 l

l 1 100 µM BPA p

p 5

R R

/ mHypoE-41

/ p

3 ####

r #### f Veh t ####

#### g #### #### A

A 10 µM BPA 0 0 100 µM BPA 0.1% 6 µM 0.1% 6 µM 0.1% 6 µM 0.1% 6 µM DMSO CC DMSO CC DMSO CC DMSO CC

64

Figure 3-5 BPA induces EndR stress, JNK, ERK and AMPK, but does not require EndR stress, JNK or ERK to increase Agrp mRNA. (A) mHypoA-59 or mHypoE-41 cells pre-treated with transcriptional inhibitor 10 µg/ml actinomycin D (ActD) or 0.1% DMSO for 1 hour, followed by treatment with 100 µM BPA or vehicle (0.05% EtOH) for 8 hours (n=4). (B) mHypoA-59 cells transfected with a siRNA duplex targeting Atf4 or negative control (NC) for 24 hours, followed by treatment with 100 µM BPA or vehicle for 8 hours (n=3). (C) mHypoA-59 or mHypoE-41 cells pre-treated with H2O, 5 mM or 10 mM sodium phenylbutryate (PBA) for 1 hour, prior to treatment with 100 µM BPA or vehicle for 8 hours. (D) mHypoA-59 and (E) mHypoE-41 cells serum starved for 1 hour and treated with vehicle, 10 or 100 µM BPA for 15 or 30 minutes. (F, H) mHypoA-59 or mHypoE-41 cells pre- treated with 0.1% DMSO or (F) 50 µM SP600125 (JNK inhibitor) or (H) 10 µM PD0352901 (MEK/ERK inhibitor) for 1 hour, followed by treatment with 100 µM BPA or vehicle for 8 hours. (G) mHypoA-59 cells serum starved for 1 hour and treated with 100 µM BPA or vehicle for 5 minutes. (I) mHypoA-59 cells treated with 100 µM BPA or vehicle for 8 hours. (J) mHypoA-59 cells treated with 500 µM 5-Aminoimidazole-4-carboxamide ribonucleotide (AICAR, AMPK activator) for 9 hours. (K, L) mHypoA-59 and mHypoE-41 cells pre-treated with 0.1% DMSO or 6 µM Compound C (CC) for 1 hour, followed by treatment with 100 µM BPA or vehicle for 8 hours. Gene expression was analyzed using qRT-PCR. Protein levels were analyzed using western blotting and representative blots from one experimental replicate are shown. Data are expressed as mean +/- SEM, and statistical significance was determined using a Two-way ANOVA, followed by the Tukey multiple comparison test (A, B, C, F, H, K, L) or by the Bonferroni post-hoc test (D,E), or using the student T-test (G,I,J); Veh vs. BPA/H2O vs AICAR: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; NC vs. siRNA/H2O/DMSO vs. inhibitor: #P<0.05, ##P<0.01, ###P<0.001, ####P<0.0001; interaction: +P<0.05, ++P<0.01, +++P<0.001, ++++P<0.0001

3.3.6 ATF3 is also involved in the BPA-mediated increase in Npy expression

Given that ATF3 plays a crucial role in regulating BPA-induced Agrp expression, we questioned whether it affects the expression of the other major orexigenic neuropeptide in the arcuate nucleus of the hypothalamus, Npy. In Chapter 5, we show that BPA-induced Npy upregulation, but not Agrp, requires the circadian regulator and transcription factor, BMAL1 (225). Knockdown of ATF3 decreased basal levels of Npy in both vehicle and BPA-treated groups in the mHypoA-59 cells (Fig 3-6A). In the mHypoE-41 cells, the BPA-induced upregulation of Npy was prevented with ATF3 knockdown (Fig 3-6A). Prevention of the Npy increase by BPA occurred with a concurrent decrease in Bmal1 expression when mHypoE-41 cells were treated with BPA in the presence of ATF3 siRNA (Fig 3-6B). These results suggest that ATF3, at least in part, also plays a role in the BPA-mediated increase in Npy expression in hypothalamic cells.

65

A mHypoA-59 mHypoE-41 + ***

A 2 ns N

R # m

* 7

l ## p

R 1 / *

y ##

p N

0 NC Atf3 NC Atf3 siRNA siRNA

B mHypoE-41

A 2

N + Legend

R mHypoA-59

m

7

l Veh

p 1 100 µM BPA R

/ **

1 mHypoE-41 l

a Veh m

B 100 µM BPA 0 NC Atf3 siRNA

Figure 3-6 ATF3 knockdown attenuates the BPA-mediated increase in Npy mRNA expression. (A) mHypoA-59 cells (n=3) or (A, B) mHypoE-41 cells (n=6 for Npy, n=3 for Bmal1) transfected with a siRNA duplex targeting Atf3 or negative control (NC) for 24 hours, followed by treatment with 100 µM BPA or vehicle for 8 hours. Npy (A) and Bmal1 (B) mRNA expression was measured using qRT-PCR. Data are expressed as mean +/- SEM, and statistical significance was determined using a Two-way ANOVA, followed by the Tukey multiple comparison test (A, B); Veh vs. BPA: *P<0.05, **P<0.01, ***P<0.001; NC vs. siRNA: #P<0.05, ##P<0.01; interaction: +P<0.05, ns = not significant 3.4 Discussion

The obesogenic effects of BPA have been described with respect to peripheral metabolism (15, 60, 64); however, how BPA affects the hypothalamic neurons that control whole-body energy homeostasis is far less studied. In fact, BPA has been shown to accumulate in the human hypothalamus (37), highlighting the importance of studying the hypothalamic effects of BPA. Here, we add to the current pool of evidence that BPA affects the hypothalamic NPY/AgRP neurons by describing that BPA increases de novo transcription of orexigenic neuropeptide Agrp

66 in hypothalamic neurons through an ATF3-dependent mechanism. We also demonstrate that BPA can alter Agrp expression in not only embryonic-derived, but also adult-derived neurons using similar mechanisms, underscoring the importance of studying the effects of BPA in adulthood. We further describe that ATF3 is also an important regulator of the BPA-mediated induction of Npy, pointing to a common mediator of orexigenic peptide inducing effects of BPA in the hypothalamus.

Perinatal exposure to BPA in rodents increases body weight (57), leads to pancreatic dysfunction (238), increases Agrp mRNA levels when combined with a high fat diet (57) and upregulates AgRP protein levels in neural progenitor cells in vivo and in vitro (184). In line with this evidence, BPA increased Agrp mRNA expression in four embryonic-derived cell lines. However, BPA increased Agrp expression in adult-derived cell lines, and importantly, adult- derived hypothalamic primary culture, suggesting that adult neurons are also susceptible to BPA- mediated dysregulation of the neuropeptides controlling energy homeostasis, albeit perhaps to a lesser degree, as both the magnitude and the temporal response of increased Agrp was limited in the adult-derived cells. Pubertal or adult exposure to BPA has also been shown to have detrimental consequences in rodent models. For instance, 5-week old rodents exposed to dietary BPA had increased body weights (60). Adult-rodent exposure to BPA also led to abnormal insulin production in the pancreas (239), insulin resistance in the muscle (63) and triglyceride accumulation in the liver (240). Furthermore, urine levels of BPA are correlated with BMI in children (48) as well as adults (46), suggesting acute, adulthood exposure to BPA is also harmful. Nevertheless, it is important to note that the timing of exposure during development or in adulthood plays a significant role in the effects of BPA; for example, perinatal exposure starting only at embryonic day 18 and subsequent postnatal exposure up to postnatal day 7 did not alter body weight nor Npy and Agrp expression in the hypothalamus of adult female rats (186). These pre-natal exposure events also prime the organism to be subsequently altered by a second hit in adulthood, such as a high fat diet (57).

BPA concentrations were initially thought to be in the low nM range in human serum and urine samples (~ 5 nM) (232). However, new measurement techniques (36), and the fact that BPA accumulates in tissues (39), has indicated that BPA concentrations may be higher than initially thought. Gerona et al. have demonstrated that older detection methods largely underestimated the levels of BPA metabolites in human urine samples, and have reported mean concentrations to be

67 approximately 230 nM, with as high as 3 M BPA detected in human urine (36). Although comparisons between urine or serum BPA concentrations and neuronal culture treatment concentrations are not entirely appropriate due to the nature of the cell culture environment, it is important to note that our treatment concentration of 100 M was much higher than the reported human concentrations. As such, the relevance of the concentration of BPA, and other hormones such as estrogen, used in in vitro studies is a primary question. As an example, circulating concentrations of E2 are in the picomolar range (19), while in vitro experiments often require 10 to 100 nM (1000-fold greater) to detect a significant effect on gene expression or function (95). This is in part due to the lipophilic nature of these compounds, which would be appropriate for the circulation and fat storage, but the inherent aqueous environment of cell culture experiments may not be conducive to easy access to the neuron. We therefore quantified how much BPA entered the neurons, with the assumption that BPA predominantly affects cells through intracellular receptor activation rather than via cell-surface receptor activation (241). One exception to this assumption is the membrane-bound ER, GPER. Although primarily characterized as residing on the endoplasmic reticulum membrane (242-244), GPER does also localize to the plasma membrane (245), and BPA has been shown to act through this receptor (189). However, in both the mHypoA-59 cells and mHypoE-41 cells, a GPER antagonist, G15, did not block the BPA-mediated induction of Agrp (Appendix A), ruling out the involvement of both plasma-membrane and ER-membrane bound GPER. Besides GPER, the majority of mechanisms through which BPA affects cells have been intracellular in nature (241). Based on our quantification, only a fraction of BPA was recovered from the cellular component of the in vitro culture the hypothalamic neurons in culture, suggesting that the exposure concentration was much less than the treatment concentration. Furthermore, the percentage decreased with lower concentrations of BPA, alluding to the need to use higher concentrations to ensure adequate cellular entry.

ATF3 is a stress-responsive transcription factor of the ATF/CREB family, and has been directly linked to the transcriptional upregulation of Agrp in response to low glucose (141). This effect occurs through binding of ATF3 to FOXO1, and then binding of both to FOXO1 responsive elements on the Agrp promoter (141). Here, we describe a FOXO1-independent effect of ATF3 since knockdown of FOXO1 did not block the increase in Agrp, and ATF3 did not show increased binding to the FOXO1 sites in the Agrp promoter. This FOXO1-independent effect

68 could either occur due to ATF3 homodimer-binding, ATF3 dimerizing with another bZIP transcription factor, followed by promoter activation, or through ATF3 altering the expression and/or binding of another transcription factor (226). The latter may be true for Npy. In chapter 5, we link the transcription factor BMAL1 to BPA-induced upregulation of Npy and show that BPA induces increased BMAL1 binding at the Npy promoter (225). The mechanism by which BPA increases BMAL1 expression and binding was not elucidated. We show here that when ATF3 is knocked down, BPA leads to a downregulation of Bmal1 expression in the mHypoE-41 cell line. Egr1 is a primary target of ATF3 (246) and is required for the daily rhythm of Bmal1 in the suprachiasmatic nucleus of mice (247). Thus, although direct promoter binding of ATF3 cannot be ruled out, it is possible that ATF3 may, in part, mediate BMAL1 binding to the Npy promoter with BPA exposure.

In terms of Agrp, either or both forms of regulation may be involved. In a study investigating the consensus sequences enriched for ATF3 binding across the human genome, Zhao et al. described that 81% of ATF3 binding sites contained the ATF:CRE, AP1 or CEBP:ATF consensus sequences (226). In fact, over half of these binding sites were in distal or enhancer regions far from gene transcriptional start sites (226). In our study, we determined whether there was increased binding of ATF3 to two putative ATF3 binding sites in the distal regulatory region of Agrp with BPA treatment. Although increased binding with BPA did not occur at either site, there was basal binding at the ATF:CRE site, as compared with IgG in the mHypoA-59 cells. Given that ATF3 knockdown increased Agrp mRNA levels in vehicle-treated mHypoA-59 cells, it is plausible that ATF3 binds and represses Agrp under non-stimulated conditions. Similarly, although Agrp expression itself was not studied, knocking out ATF3 in LepRb neurons led to increased body weight in female mice (248), suggesting ATF3 may contribute to maintaining basal energy homeostasis. With low glucose (141) or with BPA exposure, the changing transcription factor pool may result in different factors that then interact with ATF3 and change its transcriptional activity. Interestingly, Zhao et al., found that many genes whose expression was regulated by a stress-event, such as DNA damage, were pre-bound by ATF3, and other transcription factors can then be recruited to these sites (226). Thus, the idea that pre-bound ATF3 (promoter bookmarking/priming) at this site may recruit and bind to alternate transcription factors to increase transcription cannot be ruled out for Agrp. Nevertheless, depending on the binding site, stress-induced elevation of ATF3 can lead to decreased binding of ATF3 to the

69

DNA, and binding of ATF3 activates and represses an approximately equal number of genes, demonstrating the complexity in ATF3-induced regulation (226). Upon ATF3 knockdown, basal mRNA levels of Agrp and Npy were significantly altered in the mHypoA-59 cells but remained unchanged in the mHypoE-41 cells. This may be due to the presence of specific transcription factors in the adult-derived cell line that interact with ATF3 to control basal levels of Agrp and Npy expression.

Curiously, Stat3 expression was increased with BPA treatment in our experiment and BPA has been shown to induce STAT3 phosphorylation in adipocytes and macrophages (204, 249). Both Stat3 overexpression and phosphorylation would be expected to suppress Agrp expression (133, 250), suggesting that in these hypothalamic neurons, BPA increases Agrp independently of STAT3 or that the effects of other transcription factors override the effect of STAT3. In fact, although Stat3 overexpression directly increases Agrp transcription in in vitro reporter assays (133) and Stat3 knockdown is correlated with an increase in Agrp, STAT3 may not play a significant role in direct Agrp transcription in vivo as initially thought (251). The Stat3 induction in hypothalamic neurons in response to BPA may occur secondary to inflammatory cytokine signaling (203), which routinely activates STAT3 (204, 249), and occurs independently to Agrp transcription.

Other transcription factors that are involved in Agrp upregulation that we did not investigate here include the glucocorticoid receptor, brain-specific homeobox (BSX) and Krüppel-like factor 4 (KLF4). BPA can bind and activate several nuclear receptors, including the glucocorticoid receptor (64, 252). However, the glucocorticoid receptor antagonist RU486 did not block the BPA-mediated increase in Agrp in the mHypoA-59 cells (Appendix A). BSX is a transcription factor, highly expressed in the arcuate nucleus of the hypothalamus, and binds to the Agrp promoter adjacent to the FOXO1 binding sites (143, 253). It has also been shown to interact with both FOXO1 and CREB to upregulate Agrp transcription (253). KLF4 is a zinc-finger transcription factor that binds to CACCC element in the Agrp promoter (254). Interestingly, it has been described to activate transcription of Atf3 in human colorectal cancer cells (255). Whether BSX and KLF4 are altered with BPA in hypothalamic cells and whether they independently and/or with ATF3 contribute to BPA-mediated Agrp investigation warrants further investigation.

70

Alternatively, transcription factor binding sites in regions, such as the introns of downstream genes, have been implicated in the regulation of the human Agrp gene (256). Furthermore, intronic regions can enhance transcription without acting as transcription factor binding regions – a phenomenon termed intron-mediated enhancement (257). Marklund et al. showed that the human Y1 receptor undergoes such regulation, where a 97 bp intronic region leads to increased mRNA and protein expression (258). Whether the mouse Agrp gene may be regulated by intron- mediated enhancement and whether ATF3 may influence this regulation will be an interesting area of investigation.

As a stress-responsive transcription factor, ATF3 is activated by a myriad of upstream pathways and kinases. For instance, endoplasmic reticulum stress induces ATF3, along with other classic EndR stress-related factors, such as ATF4 and ATF6 (230). Belonging to the same family, these have similar consensus sequences for promoter binding. However, the DNA binding profile and gene regulation profile of ATF3 is distinct from ATF4 and ATF6, largely due to its ability to interact with a variety of other transcription factors (226). This is corroborated by our findings that although BPA induced EndR stress, ATF3, ATF4 and ATF6, the BPA-mediated increase in Agrp is only dependent on ATF3. MAPKs JNK and ERK were also activated with BPA treatment, and are known upstream regulators of ATF3 (236). However, in our study, the BPA- mediated upregulation of Agrp or Atf3 (unpublished, NL and DDB) was not blocked by JNK and ERK pathway inhibition. Interestingly, there was a non-significant trend in the mHypoA-59 cells, where the magnitude of upregulation in Agrp was decreased in the presence of the JNK inhibitor and the chemical chaperone, PBA, whereas this did not occur in the mHypoE-41 cells. Again, this may allude to the greater vulnerability of the Agrp system in embryonic neurons to BPA treatment as blocking these pathways may, in part, alleviate some of the BPA-induced alterations in the mHypoA-59 neurons.

Examining the mechanisms by which nutrient signals regulate Agrp may give clues as to how BPA may be regulating Agrp. Amino acid deprivation increases Agrp and ATF3 (259). Amino acid supplementation decreases levels of Agrp in hypothalamic neurons via mTOR activation (142). mTOR is active in states of high energy, while AMPK activation in the hypothalamus occurs during states of low energy, such as fasting. In fact, constitutively active AMPK has been shown to increase Agrp (127) and ATF3 is a downstream target of AMPK (140). We therefore evaluated whether BPA lead to AMPK activation and found that

71

BPA increases AMPK phosphorylation and thus, activation in mHypoA-59 cells. Furthermore, activation of AMPK, using AICAR, led to an increase in Agrp mRNA expression, whereas inhibition of AMPK, using compound C, inhibited the upregulation in Atf3 and Agrp. These results suggest that BPA may mimic a state of low energy in hypothalamic neurons, leading to the induction of AMPK, ATF3 and ultimately orexigenic Agrp mRNA expression. Future experiments using additional inhibitors of AMPK activity are required to confirm the relationship between AMPK, ATF3 and Agrp as AMPK can also target other factors, including BSX (260) and KLF4 (261), which can lead to the increase in Agrp.

In addition to increasing food intake, AgRP neurons also act as the main centre of actions of satiety factors insulin and leptin. Here, AgRP neurons are involved in the maintenance of whole- body glucose homeostasis by supressing gluconeogenesis through decreased vagal nerve inputs to the liver (254, 262). Given that insulin and leptin both act on AgRP neurons to mediate satiety by directly decreasing Agrp transcription, the fact that BPA leads to de novo Agrp transcription may override the effects of such satiety factors. In fact, insulin and leptin resistance have both been described with BPA exposure (183, 263).

In conclusion, we describe for the first time that BPA increases de novo transcription of Agrp through an ATF3-dependent mechanism in hypothalamic neurons. ATF3 is also, at least in part, involved in the BPA-mediated increase in Npy, suggesting a common, targetable mediator to alleviate the orexigenic neuropeptide-inducing effects of BPA in the hypothalamus.

72

Chapter 4 BPA differentially regulates Npy expression in hypothalamic neurons through a mechanism involving oxidative stress

Manuscript published in Endocrinology

Citation: Loganathan N, McIlwraith EK, Belsham DD. BPA Differentially Regulates NPY Expression in Hypothalamic Neurons Through a Mechanism Involving Oxidative Stress. Endocrinology 2020; 161, doi: 10.1210/endocr/bqaa170

Contributions: • NL completed the experiments and wrote the manuscript • EKM assisted with or completed experiments in Fig 4-1B, 4-2B and Fig 4-6 • DDB provided scientific input, direction and funding • All authors edited the manuscript

73

4 BPA differentially regulates Npy expression in hypothalamic neurons through a mechanism involving oxidative stress 4.1 Abstract

BPA, a ubiquitous endocrine-disrupting chemical, interferes with reproduction and is also considered an obesogen. The NPY neurons of the hypothalamus control both food intake and reproduction and have emerged as potential targets of BPA. These functionally diverse subpopulations of NPY neurons are differentially regulated by peripheral signals, such as estrogen and leptin. Whether BPA also differentially alters Npy expression in subpopulations of NPY neurons, contributing to BPA-induced endocrine dysfunction is unclear. We investigated the response of six immortalized hypothalamic NPY-expressing cell lines to BPA treatment. BPA upregulated Npy mRNA expression in four cell lines (mHypoA-59, mHypoE-41, mHypoA- 2/12, mHypoE-42), and downregulated Npy in two lines (mHypoE-46, mHypoE-44). This differential expression of Npy occurred concurrently with differential expression of estrogen receptor mRNA levels. Inhibition of GPER or ERß prevented the BPA-mediated decrease in Npy, wherease inhibition of energy sensor AMPK with compound C prevented BPA-induced increase in Npy. BPA also altered neuroinflammatory and oxidative stress markers in both mHypoA-59 and mHypoE-46 cell lines despite the differential regulation of Npy. Remarkably, treatment with BPA in an antioxidant-rich media, NBA, or with ROS scavenger TUDCA mitigated the BPA-induced increase and decrease in Npy. Furthermore, two anti-oxidant species from NBA, NAC and vitamin B6, diminished the induction of Npy in the mHypoA-59 cells, demonstrating these supplements can counteract BPA-induced dysregulation in certain subpopulations. Overall, these results illustrate the differential regulation of Npy by BPA in neuronal subpopulations, and point to oxidative stress as a pathway that can be targeted to block BPA-induced Npy dysregulation in hypothalamic neurons.

4.2 Introduction

NPY neurons of the hypothalamus are recognized as critical regulators of energy balance and reproduction (17). ICV injection of NPY increases food intake and chronic administration ultimately leads to obesity (264). ICV injection of NPY also blocks LH secretion in rats (100) and ewes (101), while food deprivation, which increases NPY levels, is associated with

74 decreased LH secretion (102), illustrating the inhibitory actions of NPY on the reproductive axis. However, NPY also has stimulatory actions, leading to GnRH secretion (103-105); its levels increase prior to ovulation (106); and the LH surge, which triggers ovulation, is attenuated in animals lacking NPY (107, 108). These differing effects of NPY are likely a result of activation of different receptor subtypes (109) as well as of varying hormonal environments as E2-priming of ovariectomized rats allows NPY-mediated increases in GnRH release, whereas in the absence of estradiol, NPY suppresses the reproductive axis (110). Estradiol also acts as an anorexigenic hormone, supressing food intake, partially through inhibiting Npy expression (111). As such, the feeding and reproductive functions of NPY may result from the existence of subpopulations of NPY neurons that are differentially regulated by several signals, including estradiol (17, 128). From a pathological standpoint, the link between reproduction and energy balance is clear as infertility is prevalent in individuals with abnormal energy balance (9). Given the important role of NPY in controlling food intake and reproduction, dysregulation of NPY may contribute to pathologies, such as infertility and obesity.

One factor capable of disrupting NPY regulation in the hypothalamus is the endocrine-disrupting chemical BPA. Indeed, several epidemiological and animal studies have described the detrimental effects of BPA on reproductive function and metabolic homeostasis (15). Found in plastics, canned food linings, receipt paper, and dental sealant, BPA enters the human body primarily through oral and dermal exposures (25). As a result, BPA is detected in 90-95% of tested urine samples (232, 265). Urine levels of BPA are positively correlated with polycystic ovarian syndrome (69), recurrent miscarriages (266) and reduced sperm count (72), as well as, BMI, insulin resistance and type II diabetes (47, 229, 265). Animal models exposed to BPA demonstrate altered GnRH, KISS1 (181, 267) and follicle-stimulating hormone release (267), decreased fertility (268) and increased body weight (60, 62).

Although evidence of the effects of BPA on hypothalamic reproductive peptides had been documented, whether BPA can directly affect hypothalamic control of energy balance was largely unknown until recently. McKay et al. described altered Npy and Agrp expression in animals perinatally exposed to BPA and subsequently fed a high fat diet (57). Desai et al. showed that NPY protein levels were increased in neural progenitor cells exposed to BPA (184). Furthermore, in chapter 5, we describe that acute exposure to BPA increased Npy mRNA expression in immortalized hypothalamic neurons, and this increase was dependent on circadian

75 gene and transcription factor, BMAL1 (225). Given the potent role of NPY in increasing food intake, this evidence supports the hypothesis that BPA may act as an ‘obesogen’ at the hypothalamic level.

However, as described above, NPY neurons are heterogeneous in nature, and it remains in question whether the effects of BPA on these distinct subpopulations are identical. Leptin and E2 have both been shown to differentially regulate NPY secretion or expression in subpopulations of neurons in vitro (95, 123). As BPA is considered a putative E2 mimic (15), we hypothesized that BPA will differentially affect Npy expression in distinct subpopulations of hypothalamic neurons. Thus, using clonal NPY-expressing hypothalamic cell lines, we describe that BPA upregulates Npy expression in four subpopulations, and decreases Npy in two subpopulations. These changes occurred concurrently with differential changes in estrogen receptor mRNA levels. Remarkably, we found that an anti-oxidant rich media prevented the BPA-mediated changes in Npy expression and ER mRNA expression, and ROS scavengers, N-acetylcysteine and vitamin B6 protected neurons against BPA-induced Npy upregulation. Overall, these results suggest that the detrimental effects of BPA on reproductive function and energy homeostasis may partially be mediated by its action on specific NPY neuron populations of the hypothalamus through a mechanism involving oxidative stress.

4.3 Results

4.3.1 BPA differentially alters Npy expression in subpopulations of hypothalamic neurons

Given the heterogeneous nature of NPY-expressing neurons (17, 96), we sought to determine whether BPA differentially alters Npy expression in different subpopulations of hypothalamic neurons. Therefore, four clonal cell lines derived from both embryonic and adult mice were treated with 10 to 100 µM of BPA for 4 hours. 100 µM BPA upregulated Npy expression in the mHypoA-59 and mHypoE-41 cell lines, and downregulated Npy expression in the mHypoE-46 cell line (Fig 4-1A). Notably, 25 µM BPA and 50 µM BPA downregulated Npy expression in the mHypoE-41 and mHypoE-46 cell lines, respectively (Fig 4-1A). Mean urine concentrations of BPA have been reported to be ~230 nM, with as high as 3 µM detected (36), as such our treatment concentrations are higher than human exposure. However, due to the aqueous nature of cell culture environments and the lipophilic nature of BPA, we had previously calculated the

76 percentage of BPA entry into cells by measuring the amount of BPA within the cells and the amount remaining in the media after treatment, using an enzyme-linked immunosorbent assay method. The amount of BPA entering hypothalamic cells after 4 hours and 24 hours of treatment with 100 µM BPA was approximately 10% and 18% , respectively (269). The percentage of entry was further reduced with lower concentrations of BPA, with only 2% of 10 µM BPA entering cells after 4 hours of treatment (269). As such, these hypothalamic cells are exposed to a much lower concentration than the apparent treatment concentration, and 100 µM BPA was used for further studies to ensure adequate entry of BPA into cells.

To determine the time at which these changes reach a maximum, the above-mentioned cell lines, as well as two additional, male-embryonic derived cell lines, the mHypoE-42 and mHypoE-44 cells, were treated with 100 µM BPA for 2 to 24 hours (Fig 4-1B). This analysis divided the cell lines into two different response groups: those that showed an increase in Npy mRNA with BPA exposure (mHypoA-59, mHypoE-41, mHypoA-2/12 and mHypoE-42) and those that showed a decrease (mHypoE-46 and mHypoE-44). Maximal upregulation was seen at either 8 or 16 hours across cell lines, with Npy expression demonstrating an approximately 10-fold increase in the mHypoA-59 cells at 16 hours (Fig 4-1B). This 10-fold increase at 16 hours was not present at the lower doses of 10 or 50 µM (Fig 4-1C). Maximal downregulation with BPA exposure by 50% in the mHypoE-46 and 86% in the mHypoE-44 cells occurred at 16 and 24 hours, respectively (Fig 4-1B). These results indicate that similar to estrogen exposure, distinct subpopulations of NPY neurons respond differently to BPA treatment.

77

A 4 hours 4

A **

N 3

R

m

7

l 2

p

R /

y *

p 1

N ** ** * 0 Veh 10 25 50100 Veh 10 25 50100 Veh 10 25 50100 Veh 10 25 50100 µM BPA µM BPA µM BPA µM BPA mHypoA-59 mHypoE-41 mHypoA-2/12 mHypoE-46 Cell Line B 100 µM BPA C 16 hours 12 12 10 **** 10 ****

8 8

A A

N ****

4 N

R **** 4

R

m

m

7

l 7

* ** l

p *

p

R

R

/

/ y

** y p 2 **

p 2

N N

* ******* 0 0 hours 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 Veh 10 50 100 mHypoA-59 mHypoE-41 mHypoA-2/12 mHypoE-42 mHypoE-46 mHypoE-44 µM BPA Female Male

Figure 4-1 BPA alters Npy gene expression in hypothalamic cell lines. (A) mHypoA-59, mHypoE-41, mHypoA-2/12 and mHypoE-46 cell lines treated with 10, 25, 50 or 100 µM BPA or vehicle (0.05% EtOH) for 4 hours (n=3-4). (B) Six Npy-expressing cell lines treated with 100 µM BPA or vehicle (0.05% EtOH) for 2, 4, 8, 16 and 24 hours (n=3-6). Time- matched vehicles are represented by the dotted line at y=1 and stars indicate a significant difference compared with the time-matched vehicle. (C) mHypoA-59 cells treated with 10, 50 or 100 µM BPA or vehicle (0.05% EtOH) for 16 hours (n=4). Npy gene expression was quantified using qRT- PCR. Data are expressed as mean ± SEM, and statistical significance was determined using a One- way ANOVA (A, C, each concentration compared to vehicle) or a Two-way ANOVA (B), followed by the Bonferroni post-hoc test; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

78

4.3.2 Steroid receptor antagonists, G15 and PHTPP, prevent BPA-induced decrease in Npy in the mHypoE-46 cells

Using the mHypoA-59 and mHypoE-46 cell lines as models of BPA-induced Npy upregulation and downregulation, respectively, we attempted to elucidate the mechanisms underlying BPA- mediated changes in Npy in the different subpopulations of hypothalamic neurons. We have recently described that the BPA-mediated Npy induction requires the circadian gene and transcription factor, BMAL1 (225). Nevertheless, the specific receptor that BPA binds to mediate its effect in NPY-expressing hypothalamic neurons remains unknown. As a steroid hormone mimic, BPA is described to agonize and/or antagonize several steroid hormone receptors, including GPER, ER, ER, ERR, AhR, GR and PPAR (241, 270). Several of these receptors act as transcription factors themselves, thereby directly regulating gene expression. In order to determine whether the BPA-mediated Npy changes are a result of steroid receptor agonism, cells were pre-treated with antagonists or inverse of the above-mentioned receptors (Fig 4-2A, B) or an Esr1 siRNA (Fig 4-2C, D) to reduce ER-mediated activity, followed by BPA treatment. The receptor antagonists were verified for proper functioning as described in the methods section. The receptor antagonists (Fig 4-2A) or Esr1 knockdown (Fig 4-2C) did not prevent BPA-mediated upregulation in Npy expression in the mHypoA-59 cells. Interestingly, treatment with 5 µM G15 and 5 µM PHTPP switched the BPA-induced downregulation in Npy expression in the mHypoE-46 cells to an upregulation (Fig 4-2B), but did not block the BPA-mediated increase in Npy in the mHypoA-59 cells (unpublished, NL and DDB). These results suggest that GPER and ER may be required for the BPA-mediated downregulation in Npy, but not the upregulation, as blocking these receptors allowed the mHypoE-46 cells to respond to BPA in a similar manner to the mHypoA-59 cells.

79

Legend A mHypoA-59 mHypoA-59 mHypoE-46 +++ Veh Veh 25 100 µM BPA 100 µM BPA #### +++ **** A 20 ++

N ####

R #### + **** m 15

+ **** 7

l # p

R 10 / **** **** y **** **** p *** ** ** * # N 5

0

2 2 1 6 6

P

O O O O O

5

M

M

8 8

9 8 8

% % % % %

M

P

M

S S S S S

1

µ

M

1 1

µ 1

4 4

1 1 1 1 1

M

µ

M

T

µ

. . . . .

µ

5 5

3

M M M M M

G

0

U U

µ

µ

1

H

0 0 0 0 0 1

1

2

K K

1

0

D D D D D

R R

0

1

P

2

1

S S

1

H

G G C antagonist to: GPER ERβ ERRγ AhR GR

B mHypoE-46 + +++ ## 3 #### * ** 2 ++

A #### N 1.5

R +

m

7

l 1.0 p

R # /

y * *** **** p 0.5 * *** ** N *** **

0.0

2 7

P P

5 5

O O O O O O

M M

8

0

% %

% % % %

1 1

P P

M M

µ µ

M

S S S S S S

M

1

9

1 1 1 1

T T

µ µ

5 1

n

G G

. . . .

µ

1 5

5

0

. .

M M M M M M

H H

0 0 0 0

1 5

7

0 0 0

K

0

D D D D D D

P P

0

0

1

S

0

1

G T antagonist to: GPER ERβ ERRγ PPARγ

C mHypoA-59 D mHypoE-46

i 1.5 ii 4 i 1.5 ii

++ 1.5

A A

++++ A **

N

A

N N

3 N

R R R ** 1.0 R

1.0 m

m m

1.0

m

7

7

l 7

* l l

2 7

p

l

p

p

p

R

R

/ R

/ **

R /

#### / 1

0.5 1 0.5 y 0.5

r * y

### r ****

p

s p

#### 1 s

E

N

N E

0.0 0 0.0 0.0 NC Esr1 NC Esr1 NC Esr1 NC Esr1 siRNA siRNA siRNA siRNA

80

Figure 4-2 Steroid receptor antagonism does not abolish effect of BPA on Npy upregulation, but antagonism of GPER or ER reverses BPA-mediated downregulation of Npy expression in mHypoE-46 cells. (A) mHypoA-59 cells pre-treated with indicated concentrations of antagonist or 0.1 % DMSO for 1 hour prior to treatment to 100 µM BPA or vehicle (0.05% EtOH) for 8 hours or 16 hours (for PHTPP experiment) (n=3-4). (B) mHypoE-46 cells pre-treated for 1 hour with antagonists or 0.1%/0.5% DMSO, followed by treatment with 100 µM BPA or vehicle (0.05% EtOH) for 16 hours (n=3-5). (C) mHypoA-59 and (D) mHypoE-46 cells were transfected with an siRNA targeting Esr1 for 24 hours, followed by treatment with 100 µM BPA or vehicle for 8 or 16 hours, respectively (n=3-4). siRNA knockdown of Esr1 mRNA expression was confirmed (Ci, Di) prior to analyzing changes in Npy expression by qRT-PCR. Data are presented as mean ± SEM, and statistical significance was determined using a Two-way ANOVA, followed by the Tukey multiple comparison test; Veh vs. BPA: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; DMSO vs. antagonist or NC vs. Esr1 siRNA: #P<0.05, ##P<0.01, ###P<0.001, ####P<0.0001; interaction: +P<0.05, ++P<0.01, +++P<0.001, ++++P<0.0001.

4.3.3 Estrogen receptor mRNA levels are differentially regulated with BPA treatment

Besides acting as an agonist of steroid hormone receptors, BPA has been shown to directly alter the expression levels of ERs, Esr1 and Esr2, in hypothalamic POMC-expressing neurons (203). Additionally, the relative levels of ERs are crucial for the regulation of Npy expression, as described by Titolo et al., who demonstrated that the differential regulation of Npy expression by estradiol in two subpopulations of hypothalamic neurons was attributed to the ratio of estrogen receptors present (95). Thus, we investigated potential subpopulation-specific differences in BPA-mediated changes in Esr1 and Esr2 mRNA levels. In the mHypoA-59 cells, 100 M BPA treatment decreased Esr2 mRNA levels from 2 to 24 hours, while Esr1 mRNA levels were only modestly reduced at 24 hours (Fig 4-3Ai). In contrast, in the mHypoE-46 cells, Esr2 mRNA levels were decreased from 2 to 24 hours and Esr1 mRNA levels showed a greater decrease from 8 to 24 hours compared with Esr2 (Fig 4-3Ai). As a result, the ratio of Esr2/Esr1 was decreased with BPA treatment in the mHypoA-59 cells and increased with BPA treatment in the mHypoE- 46 cells relative to vehicle treatment (Fig 4-3Aii). This phenomenon of a greater decrease in Esr2 mRNA levels compared with Esr1 mRNA levels in the mHypoA-59 cells with BPA exposure is replicated in the other three cell lines that show an increase in Npy expression with 100 M BPA treatment (mHypoE-41, mHypoE-42, mHypoA-2/12; Esr2:Esr1 ratio <1; Figure 4-3B). Esr1 expression was very low in the mHypoE-44 cells (Table 4-1). Thus, Esr1 mRNA levels are lower in the two cell lines showing a decrease in Npy with 100 M BPA treatment compared with the other 4 cell lines (Fig 4-3B, Fig 4-1B). qPCR primer efficiencies of Esr1 (~94%) and

81

Esr2 (~106%) primers were established using standard curves conducted with cDNA from the mHypoE-41 cell line as well as the mHypoA-2/12 cell line. Basal expression levels of estrogen receptors and other transcriptional regulators are described in Table 4-1.

Table 4-1 Expression levels (Ct) of transcriptional regulators in the six NPY-expressing cell lines

Gene Name Cell Line (mHypoX-XX)

A-59 E-41 A-2/12 E-42 E-46 E-44

Esr1 26.1 28.0 26.3 29.9 28.1 32.6

Esr2 27.2 26.0 28.9 28.3 27.4 28.4

Tfap2b 23.5 - 21.7 - - -

Hspb1 23.4 25.7 23.5 25.4 26.9 28.4

Ar 22.3 26.3 21.4 26.9 26.1 26.6

Ncr3c1 21.1 21.5 20.5 21.2 21.2 21.1

Ppar훾 24.4 24.1 25.2 23.7 24.4 24.9

Rpl7 18.0 18.5 17.4 17.7 17.7 17.3

Values represent mean cycle at threshold (Ct) levels as determined by qRT-PCR.

As Esr2 was downregulated in both mHypoA-59 cells and mHypoE-46 cells, and a larger Esr1 downregulation was unique to mHypoE-46 cells, we questioned whether the lack of Esr1 may mediate the changes in Npy in mHypoE-46 cells. However, overexpression of Esr1 alone was not sufficient to prevent the BPA-induced downregulation of Npy in the mHypoE-46 cells (Fig 4- 3C).

82

A i ii

1.5 2.5

o

i

Esr1 (ERα) Esr1 (ERα) t A

a **** r

Esr2 (ERβ) Esr2 (ERβ)

N 2.0

A R

N *

m 1.0

vehicle y=1

R 1.5

7

l

m

p

1

R r / *** 1.0

** **** s

e 0.5 **** **** E

n **** **** **** / **** e **** **** * **** 2 ****

r 0.5 **** G

*** s

**** E 0.0 *** 0.0 hours 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 mHypoA-59 mHypoE-46 mHypoA-59 mHypoE-46 B 1.5

Esr1 (ERα)

A Esr2 (ERβ) N

R vehicle y=1

m 1.0

7

l

p R / ** e 0.5 **** n **** e ** * ** G *** *** ** 0.0 **** ** **** hours 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 Esr2:Esr10.91 0.79 0.95 0.87 0.61 0.97 0.76 0.73 0.53 0.32 0.78 0.62 0.87 0.70 0.60 mHypoE-41 mHypoA-2/12 mHypoE-42 mHypoE-44

C mHypoE-46 - ERα overexpression

i ## ii 160 ## Veh

100 µM BPA A

120 A

N

N R

80 R 1

m

m

7

2.0 7

l

l p

1.5 p

R

R /

/ ** y

1 ** r

1.0 p

s N E 0.5 0.0 0 Control ERα + Control ERα + 16 hours 16 hours

Figure 4-3 BPA differentially alters estrogen receptor mRNA levels in hypothalamic cell lines. (A, B) Cells treated with 100 µM BPA or vehicle (0.05% EtOH) for 2, 4 8, 16 and 24 hours (n=3- 4). Open circles represent the level of Esr1 mRNA and closed squares represent the level of Esr2 mRNA compared to the time-matched vehicle indicated by the dotted line at y=1. Stars denote statistical significance of each gene compared with its time-matched vehicle control. The magnitude change in Esr2 over Esr1 with BPA exposure is represented by a graph (Aii) for the mHypoA-59 and mHypoE-44 cell lines, and by a numerical value below each graph in B. Esr1

83 was not quantified in the mHypoE-44 cells due to low expression and a resulting high degree of variability. (C) mHypoE-46 cells transfected with a control (pcDNA3.1+) or an Esr1 (ER) overexpression plasmid for 6 hours, followed by treatment with 100 µM BPA or vehicle (0.05% EtOH) for 16 hours. Gene expression was analyzed using qRT-PCR. Data are expressed as mean ± SEM, and statistical significance was determined using a Two-way ANOVA, followed by the Bonferroni post-hoc test (A, B) or the Tukey multiple comparison test (C); Veh vs. BPA: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; Control vs. ER+ plasmid: ##P<0.01.

4.3.4 AMPK inhibition prevents BPA-mediated upregulation in Npy expression

Next, we questioned whether a common upstream pathway is differentially regulated by BPA in the two cell lines, leading to the differential changes in Npy and estrogen receptors. AMPK activation in the hypothalamus occurs during states of low energy, such as fasting. Inhibition of hypothalamic AMPK was shown to decrease Npy expression in ad-lib fed animals, while constitutively active AMPK elevated fasting-induced Npy, demonstrating the involvement of AMPK in Npy expression (127). Interestingly, activation of AMPK using AICAR decreased the Esr2/Esr1 mRNA ratio in both the mHypoA-59 cells and mHypoE-46 cells (Fig 4-4A). Therefore, we examined whether BPA differentially alters AMPK activity in the mHypoA-59 and mHypoE-46 cell lines, thereby contributing to differential regulation of the estrogen receptors and Npy. We have previously described that BPA treatment led to a 2-fold upregulation in pAMPK levels in the mHypoA-59 cells (269). However, BPA also increased pAMPK levels in the mHypoE-46 cells at 16 hours (Fig 4-4B), suggesting the differential regulation of Npy in the two cell lines does not originate from differences in AMPK activation. In line with this, inhibition of AMPK by CC blocked the BPA-mediated increase in Npy expression and altered the Esr2/Esr1 mRNA ratio in the mHypoA-59 cells (Fig 4-4C). CC decreased standardization gene Rpl7 mRNA levels in the mHypoE-46 cells. However, preliminary un-normalized results showed that the decrease in Npy in the mHypoE-46 cells is not blocked by CC (unpublished, NL and DDB). Thus, AMPK activation is required for the BPA-mediated increase in Npy expression in the mHypoA-59 cells, but likely not the decrease in the mHypoE-46 cells. Mechanisms underlying the distinct regulation of Npy by BPA between the two cell populations must therefore involve molecules downstream of or independent of AMPK.

84

A AICAR B mHypoE-46 C mHypoA-59 Veh

i ++++ ii 100 µM BPA o

i 5

t

o

K

i a

* **** t r

P ++

a

A

r

M

A 4

N

A

A

N

1 1 R

l N

R *** 1

a

m t

* R

3

m

o

7

t

l

m

/

1

p

r

K

1

s r R 2

P ####

/

s

E

/

y

M

E

2

/

p r

A ####

2 N

s 1

p

r

s E

0 0 E H O AIC H O AIC Veh 100 µM 0 0 2 2 0.1% 6 µM 0.1% 6 µM BPA A-59 E-46 DMSO CC DMSO CC pAMPK

total AMPK Figure 4-4 Inhibition of AMPK blocks BPA-mediated increase in Npy in mHypoA-59 cells. (A) mHypoA-59 (n=4) and mHypoE-46 (n=3) cells treated with 500 µM AICAR (AIC) or vehicle (H2O) for 9 or 17 hours. (B) mHypoE-46 (n=6) cells treated with 100 µM BPA or vehicle (0.05% EtOH) for 16 hours. Levels of pAMPK (Thr172) and total AMPK were determined using Western Blot analysis and representative blots from one experimental replicate are shown. (C) mHypoA-59 cells pre-treated with 6 µM Compound C (CC) for 1 hour, followed by treatment with 100 µM BPA or vehicle (0.05% EtOH) for 8 hours. Gene expression was analyzed using qRT- PCR and data are expressed as mean ± SEM. Statistical significance was determined using a student T-test (A, B) or a Two-way ANOVA, followed by the Tukey multiple comparison test (C); Veh vs. BPA or H2O vs AIC: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; DMSO vs. CC: ####P<0.0001; interaction: ++P<0.01, ++++P<0.0001.

4.3.5 BPA induces oxidative stress and neuroinflammation in Npy- expressing hypothalamic neurons

Since BPA activated AMPK in both cell lines despite differential Npy regulation, we predicted BPA may have similar upstream actions in both subpopulations that can be targeted to prevent the neuropeptide dysregulation. AMPK phosphorylation has been previously shown to occur aberrantly in response to increased oxidative stress and inflammation (218). As BPA induces oxidative stress and inflammatory cytokines in other cell types (203, 211), we investigated changes in genes related to oxidative stress and neuroinflammation in the mHypoA-59 and mHypoE-46 cell lines. Expression of Nos2, the gene encoding inducible nitric oxide synthase (iNOS), which contributes to elevated reactive nitrogen species, is increased approximately 8 to 10-fold from 4 to 16 hours in the mHypoA-59 and at 8 hours in the mHypoE-46 cells, although

85 to a lesser degree (Fig 4-5Ai). mRNA levels of anti-oxidant enzymes superoxide dismutase 1 (Sod1) and glutathione peroxidase 1 (Gpx1) were both downregulated at 16 and 24 hours in the mHypoA-59 cells and at 16 hours in the mHypoE-46 cells (Fig 4-5Aii and iii), while catalase mRNA levels were only upregulated in the mHypoA-59 cells at 24 hours (Fig 4-5Aiv). Expression of nuclear factor erythroid 2-related factor 2 (Nrf2), a major transcription factor involved in the anti-oxidant response to oxidative stress (271), remained unchanged in both cell lines, but expression of the Nrf2-target gene heme oxygenase 1 (Hmox1) was increased 4 to 8- fold in both cell lines, indicative of increased oxidative stress and increased transcriptional activity of Nrf2 in response to BPA (Fig 4-5Av and vi). Furthermore, BPA treatment upregulated markers of neuroinflammation, including tumor necrosis factor  (Tnf) and interleukin 6 (Il6) (Fig 4-5Bi – ii) in both cell lines. Expression of anti-inflammatory gene Il10 was also upregulated in the mHypoA-59 cells at 4 and 8 hours (Fig 4-5Biii), albeit to a lesser degree than the pro-inflammatory Tnf and Il6.

86

A Oxidative Stress B Neuroinflammation

i 15 iv 2.0 i 6

****

*** A

A ****

A N

N *** N

*** 1.5 R

R R ** m

10 4

m

m

7

l

7

l

7

p l

p 1.0 ** *

p

R

/

R

/

R

α

/

f 2

5 t 2

s

n a

o 0.5

T C N *** 0 0.0 0 hours 2 4 8 16 24 8 16 hours 2 4 8 16 24 8 16 hours 2 4 8 16 24 8 16 mHypoA-59 mHypoE-46 mHypoA-59 mHypoE-46 mHypoA-59 mHypoE-46

ii 1.5 v 1.5 ii 6

****

A

A

A

N

N

N R R ** ***

1.0 1.0 R 4

m

m

m

7

l 7

* l 7

p ** **

l p

p *

R

R

/

/

R

1 /

0.5 2 0.5 2

f

d

6

r

l

o

I

N S 0.0 0.0 0 hours 2 4 8 16 24 8 16 hours 2 4 8 16 24 8 16 hours 2 4 8 16 24 8 16 mHypoA-59 mHypoE-46 mHypoA-59 mHypoE-46 mHypoA-59 mHypoE-46

iii 1.5 vi 10 iii 6

A

A ****

N A

N 8 ****

R N

R ****

R m

****

m 1.0 4

m

7 l

6

7

l

p 7

* l

p R *** p

/ *

R

/ R 1 4 ****

/ ** ** 1

0.5 x 2

0

x o

** 1

p

l I

m 2

G H 0.0 0 0 hours 2 4 8 16 24 8 16 hours 2 4 8 16 24 8 16 hours 2 4 8 16 24 8 16 mHypoA-59 mHypoE-46 mHypoA-59 mHypoE-46 mHypoA-59 mHypoE-46

Figure 4-5 BPA induces oxidative stress and neuroinflammation in NPY-expressing hypothalamic cell lines. (A) mHypoA-59 and (B) mHypoE-46 cells treated with 100 µM BPA or vehicle (0.05% EtOH) for 2, 4 8, 16 and 24 hours (mHypoA-59, n=3-4) or 8 and 16 hours (mHypoE-46, n=3-4). Time- matched vehicles are represented by the dotted line at y=1 and stars indicate a significant difference compared with the time-matched vehicle. Gene expression was analyzed using qRT-PCR. Data are expressed as mean ± SEM, and statistical significance was determined using a Two-way ANOVA, followed by the Bonferroni post-hoc test; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

87

4.3.6 Antioxidant-rich media, Neurobasal A, mitigates BPA-induced changes in Npy and oxidative stress genes

We therefore examined whether the BPA-mediated changes in Npy expression can be blocked by inhibiting either oxidative stress or neuroinflammation. Neurobasal A (NBA) media is often used to cultivate primary neuronal cells due to its enhanced anti-oxidant properties compared to DMEM (272). Treatment of the mHypoA-59 cells with 100 µM BPA in the anti-oxidant rich NBA media mitigated the 10-fold upregulation in Npy seen at 16 hours in DMEM (Fig 4-6Ai). NBA downregulated Sod1 levels, however, a further decrease with BPA was prevented in the presence of NBA media (Fig 4-6Aii). Furthermore, the BPA-mediated decrease in Gpx1 was reversed with NBA media (Fig 4-6Aiii), suggesting increased synthesis of this anti-oxidant enzyme. Likewise, the BPA-induced downregulation of Npy, Sod1 and Gpx1 in the mHypoE-46 cells was abolished in the presence of NBA (Fig 4-6B). Thus, reducing BPA-induced oxidative stress using NBA protects both mHypoA-59 and mHypoE-46 cells, despite their differential responses in BPA-induced Npy expression, suggesting that oxidative stress may trigger differential transcriptional pathways in the two subpopulations.

Additional scavengers of ROS were used to confirm if BPA-mediated oxidative stress indeed leads to the changes in Npy expression. Tauroursodeoxycholic acid (TUDCA) is a common ROS scavenger, which is proposed to inhibit ROS generation by blocking the MAPK JNK (273) and/or by increasing mitochondrial biogenesis (274). Furthermore, nordihydroguaiaretic acid (NDGA) is proven to potently scavenge ROS by directly interacting with free radicals (275) and by inhibiting lipoxygenases (276). Remarkably, pre-treatment with TUDCA, the JNK inhibitor SP600125 (SP) or NDGA reduced the BPA-mediated increase in Npy expression in the mHypoA-59 cells (Fig 4-6C). While TUDCA and SP prevented a decrease in Npy in the mHypoE-46 cells, with TUDCA leading to a reversal of the response, NDGA pre-treatment did not block the ability of BPA to decrease Npy (Fig 4-6D). In contrast, inhibition of the inflammatory NFκB pathway, using PS1145 (an IKK inhibitor), only partially mitigated the increase in Npy in the mHypoA-59 cells, and did not block the decrease in Npy in mHypoE-46 cells (Fig 4-6C and 4-6D). Overall, these results suggest that while differential mechanisms may be involved downstream, oxidative stress plays an important role in BPA-mediated changes in Npy expression.

88

A mHypoA-59

i 15 ++++ ii iii

++ ++

A A

A ++++

N N

N **** R

R # R

10 1 m 1

m

m

7 7

* l

l # 7

l ####

### p

p

p R

*** R

/ /

R ****

/ *

1 1

y 5

#### x

# d

p

p

o

N

S G 0 0 0 DMEM NBA DMEM NBA DMEM NBA DMEM NBA DMEM NBA DMEM NBA 8 hours 16 hours 8 hours 16 hours 8 hours 16 hours B mHypoE-46 ++++ + i ii iii ####

++++ A

A **

A

N N

N ###

R R R ###

1 1 m 1

m

m

##

7

7

l

l

7

l

p

p

p

R R

**** /

/

R

/

1

1 y

x *** d

p ***

p

o

N

S G 0 0 0 DMEM NBA DMEM NBA DMEM NBA 16 hours 16 hours 16 hours C mHypoA-59 +++ ++++ **** A 10 ++++

N ****

R m

**** 7

l +

p 5 R

/ ####

y ### **** # p * #### N ***

0 H2O 2.5 mM 0.1% 50 µM 0.1% 4 µM 0.2% 20 µM TUDCA DMSO SP DMSO NDGA DMSO PS1145 D mHypoE-46 +++

## 2.5 **

A 2.0 N

R Legend

m #

1.5 + mHypoA-59

7 ##

l Veh p

R 1.0 *** 100 µM BPA / y mHypoE-46 p * ## *** *** N 0.5 ### Veh 100 µM BPA 0.0 H2O 2.5 mM 0.1% 50 µM 0.1% 4 µM 0.2% 20 µM TUDCA DMSO SP DMSO NDGA DMSO PS1145

89

Figure 4-6 Treatment with BPA in Neurobasal A media or with ROS scavengers protects cells from Npy mRNA dysregulation. (A) mHypoA-59 and (B) mHypoE-46 cells treated with 100 µM BPA or vehicle (0.05% EtOH) for 8 and 16 hours (mHypoA-59 cells, n=4) or 16 hours (mHypoE-46 cells, n=4) in phenol-red free Dulbecco’s Modified Eagles Medium (DMEM) or in phenol-red free Neurobasal A medium (NBA). (C, D) Cells pre-treated with 2.5 mM tauroursodeoxycholic acid (TUDCA), 50 µM SP600125 (SP), 4 µM nordihydroguaiaretic acid (NDGA) or 20 µM PS1145 or their respective vehicles for 1 hour, followed by treatment with 100 µM BPA or vehicle (0.05% EtOH) for 8 hours (C, mHypoA-59, n=3-5) or 16 hours (D, mHypoE-46, n=3-5). mRNA levels of Npy, Sod1 and Gpx1 were measured using qRT-PCR. Data are presented as mean ± SEM. Statistical significance was determined using a Two-way ANOVA, followed by the Tukey multiple comparison test; Veh vs. BPA: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; DMEM vs. NBA or DMSO vs. pre- treatments: #P<0.05, ##P<0.01, ###P<0.001, ####P<0.0001; interaction: +P<0.05, ++P<0.01, +++P<0.001, ++++P<0.0001.

4.3.7 NAC and vitamin B6 mitigate the BPA-mediated increase in Npy expression

The specific components in NBA that protected the cells from BPA-mediated Npy dysregulation are of interest as they are likely supplements, rather than chemical inhibitors, with fewer side effects. A comparison of NBA and DMEM revealed three major anti-oxidant advantages to

NBA: increased bio-availability of L-cysteine and vitamin B6 and the presence of vitamin B12 (Fig 4-7A). Pre-treatment with NAC for 16 hours blocked the effect of BPA on Npy expression in the mHypoA-59 cells (Fig 4-7Bi). Furthermore, the BPA-mediated increase in Npy expression at both 8 and 16 hours in the mHypoA-59 cells was mitigated by co-treatment with 20 or 40 µM

Vitamin B6 (in the form of pyridoxal HCl) (Fig 4-7Bii), while there was no preventative effect with vitamin B12 co-treatment (Fig 4-7Biii). NAC and the vitamins did not prevent the decrease in Npy in the mHypoE-46 cells (Fig 4-7C). Thus, cysteine and vitamin B6 may mediate the protective effect NBA has against BPA-mediated increases in Npy expression.

90

A Formulation Cysteine Vitamin B6 Vitamin B12 Legend DMEM L-Cystine HCl 20 µM pyrodoxine HCl - mHypoA-59 mHypoE-46 Veh Veh NBA L-Cysteine 20 µM pyridoxal HCl 5 nM 100 µM BPA 100 µM BPA

B mHypoA-59 +++ i ii + iii + +++ 5 +++ 10 **** 15

##

A A

4 **** 8 +++ A ###

N N

+ # N **** R

R ****

**** R 10

m

m

m

3 6

7 7

**** 7

l

l

l

p p

*** p *** ***

R R

2 4 #### R

/

/

/

y y

### #### y 5

p

p

p

N N 1 2 N

0 0 0 H2O 10 mM H2O 20 µM 40 µM H2O 20 µM 40 µM H2O 5 100 500 NAC Vitamin B6 Vitamin B6 nM Vitamin B12 8 hours 8 hours 16 hours 16 hours C mHypoE-46 i ii iii 1.5 1.5 1.5 +

+ +

A

A

A

N

N

N

R R

1.0 1.0 R 1.0

m

m

m

# ##

7

l

7

7

l

l

p

p

p

R

R

R

/

/ /

y 0.5 **** y

**** 0.5 y 0.5

p p p **** ****

N *** ** *** ****

N N

0.0 0.0 0.0 H O 10 mM 2 H2O 20 µM 40 µM H2O 5 nM 500 nM NAC Vitamin B6 Vitamin B12 16 hours 16 hours 16 hours

Figure 4-7 N-acetylcysteine and vitamin B6 mitigate BPA-induced Npy upregulation in mHypoA-59 cells. (A) Differences in the anti-oxidant components of phenol-red free DMEM (Hyclone, cat. #SH3028401) and phenol-red free NBA (Gibco, cat. #12349015). (B) mHypoA-59 cells pre- treated with 10 mM NAC for 16 hours (Bi, n=3), co-treated with 20 or 40 µM vitamin B6 (Bii, n=4-6) or 5 nM, 100 nM or 500 nM vitamin B12 (Biii, n=3-5) alongside vehicle (H2O) control, and treated with 100 µM BPA or vehicle (0.05% EtOH) for 8 or 16 hours (as indicated). (C) mHypoE- 46 cells pre-treated with 10 mM NAC for 8 hours (Ci, n=4), co-treated with vitamin B6 (Cii, n=3) or vitamin B12 (Ciii, n=3) or vehicle (H2O), and treated with 100 µM BPA or vehicle (0.05% EtOH) for 16 hours. mRNA levels of Npy were measured by qRT-PCR. Data are presented as mean ± SEM. Statistical significance was determined using a Two-way ANOVA, followed by the Tukey multiple comparison test; Veh vs. BPA: **P<0.01, ***P<0.001, ****P<0.0001; H2O vs. NAC/vitamins: #P<0.05, ##P<0.01, ###P<0.001, ####P<0.0001; interaction: +P<0.05, +++P<0.001.

A mHypoA-59 i ii

1.5 ## 1.5 ++++ A

A ###

N

N R R **

1.0 1.0

m

m

7

7

l

l

p

p R

R ##

/

/ 2

1 0.5 0.5

r

r s

s *** E E 91 0.0 0.0 DMEM NBA DMEM NBA 8 hours 8 hours 4.3.8 Dysregulation of estrogen receptor mRNA levels is reversed in anti- oxidant rich media B mHypoA-59 i ii 1.5 + 1.5 +

Since NBA protected cells from BPA-mediated Npy dysregulation,+ we questioned whether+ the A

A #

N

N R concomitant dysregulation in estrogen receptor mRNAR levels was also blocked. Remarkably, the

1.0 1.0 m

m #

#

7

7 l

downregulation of Esr2 in the mHypoA-59 cells andl both Esr1 and Esr2 in the mHypoE-46 cells

p

p

R

R /

/ **** 2

1 0.5 0.5 r

were abolished in the anti-oxidant rich NBA mediar (Fig 4-8A and 4-8C), suggesting that

s

s

E E oxidative stress may mediate changes in the estrogen receptors and Npy. Likewise, vitamin B6 0.0 0.0 H O 20 µM 40 µM co-treatment, in addition to mitigating the increase in NpyH2,O partially20 µM 40prevented µM the 2 Vitamin B6 Vitamin B6 downregulation of Esr2 in the mHypoA-59 cells at 8 hours (Fig8 hour 8B).s 8 hours

A mHypoA-59 C mHypoE-46 ++++ i ii i ii 1.5 ## 1.5 ++++ ####

2 2 **** A

A ###

A

A

N

N

N

N R

R ** R

1.0 1.0 R m

m ++++

m

m

7

7

l

l

7

7

l

p l

p 1 1 p

p #### R

R ## ####

/

/

R

R

/

2 /

1 0.5 0.5

r

r

2

1

s

s r

*** r

s

s

E

E E E **** 0.0 0.0 **** DMEM NBA DMEM NBA 0 0 DMEM NBA DMEM NBA 8 hours 8 hours 16 hours 16 hours B mHypoA-59 Legend i ii mHypoA-59 mHypoE-46 1.5 + 1.5 + Veh Veh

+ + 100 µM BPA 100 µM BPA A

A #

N

N R

R 1.0 1.0 m

m #

#

7

7

l

l

p

p

R

R /

/ **** 2

1 0.5 0.5 r

r

s

s

E E 0.0 0.0 H2O 20 µM 40 µM H2O 20 µM 40 µM Vitamin B6 Vitamin B6 8 hours 8 hours

C mFigureHypoE -46-8 Neurobasal A media++++ and vitamin B6 mitigate BPA-induced changes in Esr1 and Esr2 gene expression. i ii ####

(A,2 B) mHypoA-59 cells treated2 with 100**** µM BPA or vehicle (0.05% EtOH) for 8 hours in DMEM A

A or NBA media (A, n=4) or in the presence of vitamin B6 (B, n=4). (C) mHypoE-46 cells treated

N

N R

R with 100++++ µM BPA or vehicle (0.05% EtOH) for 16 hours in DMEM or NBA media (n=4). mRNA

m

m

7 7

expression of Esr1 and Esr2l were measured with qRT-PCR. Data are presented as mean ± SEM.

l 1 1 p

p #### #### R

R Statistical significance was determined using a Two-way ANOVA, followed by the Tukey

/

/

2

1

r

r s

s multiple comparison test; Veh vs. BPA: **P<0.01, ***P<0.001, ****P<0.0001; DMEM vs. NBA E E **** or H2O**** vs. vitamin B6: #P<0.05, ##P<0.01, ###P<0.001, ####P<0.0001; interaction: +P<0.05, ++++0 P<0.0001. 0 DMEM NBA DMEM NBA 16 hours 16 hours

Legend mHypoA-59 mHypoE-46 Veh Veh 100 µM BPA 100 µM BPA 92

4.4 Discussion

The regulation of NPY by feeding-related signals and reproductive hormones and peptides is essential for hypothalamic control of energy homeostasis and reproduction. Insulin (131), leptin (123) and estrogen (95, 111) act partially through NPY neurons, by downregulating NPY expression or secretion, to mediate states of satiety. NPY also both positively and negatively regulates GnRH neuron activation, GnRH expression and LH secretion depending on the steroid environment (110) and NPY receptor subtype activated (109). As evidenced by this dual role of NPY neurons in regulating energy balance and reproduction, the NPY neuronal population is heterogeneous in nature, which has been described both in vivo and in vitro (95, 96, 277). As such, while leptin and estrogen suppress NPY expression and secretion in specific subpopulations, leptin, E2 and KISS1 increase NPY expression or secretion in other subpopulations (95, 123, 278). The hypothalamic NPY system has previously been shown to be affected by both pre-natal and acute exposure to the environmental chemical BPA (57, 184, 225). Here, we describe that specific subpopulations of hypothalamic NPY-expressing cell lines respond differentially to acute BPA treatment. Furthermore, we demonstrate that the differential changes in Npy expression occur through the induction of oxidative stress, despite the potential involvement of distinct downstream mechanisms. While AMPK activation is necessary for BPA- mediated increases in Npy expression, steroid receptors, GPER and ERß, likely mediate the downregulation of Npy in certain subpopulations.

Several lines of evidence suggest that BPA acts as an estrogen mimic. Whether this involves ER agonism or antagonism depends on the tissue type and concentration of BPA used in the study (270). The cells in our study were exposed to 100 µM BPA although only 10 – 18% entered the cells after 4 and 24 hours (10 and 18 µM) (269). These concentrations are higher than the recently reported average concentrations of 230 nM and the maximum urine concentrations of 3 M (36), and whether neurons are exposed to such concentrations remains controversial. However, using 1000-fold higher than the circulating concentrations of steroid hormones is a common practice with in vitro experiments to observe functional effects due to the aqueous environment of cell culture (19, 95). Furthermore, BPA has been detected in human brain tissue (37), suggesting its potential to accumulate over time. We have previously shown estradiol-induced downregulation of Npy expression in both the mHypoE-42 (95) and mHypoA- 59 cells (93). As BPA upregulated Npy expression in both of these cell lines at the same time

93 point, estrogen and BPA likely have different mechanisms of action in these subpopulations. This is corroborated by the finding that estrogen receptor antagonists (G15 and PHTPP) or siRNA knockdown of Esr1 did not abolish BPA-induced Npy in the mHypoA-59 cells. Meanwhile, the fact that G15 and PHTPP switched the BPA-mediated downregulation in Npy in mHypoE-46 cells to an upregulation indicates that activation of both GPER and ERß are required for the downregulation. This likely reflects the estrogen-like action of BPA in the mHypoE-46 cells. In the absence of this activity, BPA acts on the mHypoE-46 cells as it would act on the mHypoA-59 cells, suggesting that the transcription factors specific to the mHypoA-59 cells may prevent activation of GPER and/or ERß by BPA.

BPA mediated changes in Npy expression occurred concurrently with changes in estrogen receptor mRNA expression. The ER subtype (ERα versus ERß) dominant in a cell can differentially modulate the effects of estradiol (95) or BPA (279). As an example of this, Titolo et al. showed that estradiol-induced Npy upregulation was dependent on the relative increase in Esr2 and ERß compared to ERα, while both Esr2 and Esr1 were reduced when estradiol downregulated Npy expression (95). In contrast, with BPA, the magnitude of Esr2 downregulation was consistently greater than Esr1 downregulation in cell lines showing increased Npy expression. While relative levels of ERß to ERα did mediate estradiol-induced changes in Npy expression (95), overexpression and siRNA knockdown of Esr1 alongside BPA treatment to modulate the relative levels of the receptors did not prevent the population-specific BPA-mediated changes in Npy expression. This suggests that the differential changes of Npy and the receptors are not dependent on each other, but may be modulated by a common upstream pathway.

Esr1 and Esr2 mRNA levels have been previously shown to change in the hypothalamus with pre-natal exposure to BPA; however, the direction varied depending on the age or sex of animals, concentration of BPA used and the hypothalamic nuclei studied (280, 281). This suggests that the specific transcription factors and/or co-regulators present at different stages of development, activated with differing BPA concentrations, or expressed in specific hypothalamic nuclei likely mediate BPA-induced regulation of the ERs. In our study, although the mHypoA-59 cells were derived from an adult female mouse and the mHypoE-46 cells were derived from an embryonic male mouse, these sex differences are not likely to be the sole mediator of the differential effect since BPA-mediated increases in Npy occurs in adult and embryonic male-

94 derived cell lines as well (mHypoA-2/12, mHypoE-42). Overall, as estradiol is an anorexigen, and Esr1 knockout models are obese (282), BPA may act as an orexigen in hypothalamic neurons by not only increasing Npy levels, but also by decreasing estrogen receptor levels.

In spite of distinct effects on Npy and Esr1/2 mRNA levels, BPA increased AMPK phosphorylation in both the mHypoA-59 (269) and mHypoE-46 cells. This commonality led us to further investigate whether BPA activates identical upstream pathways in both subpopulations, which can be blocked to protect both cells from Npy dysregulation. Activating AMPK, however, is known to increase Npy levels (127) and is shown here to reduce the Esr2/Esr1 ratio in both subpopulations. Thus, although AMPK is involved in the BPA-induced upregulation of Npy in the mHypoA-59 cells (as seen with CC inhibition of AMPK), it is likely a by-product of an upstream signaling cascade and does not mediate changes in Npy in the mHypoE-46 cells (Fig 9).

AMPK activity is increased in response to a low ATP:ADP ratio or low energy status. However, several factors can abnormally activate AMPK levels, one of which is oxidative stress (218). Under physiological conditions, ROS signal to activate POMC neurons and inhibit NPY neurons in states of satiety. Excessive ROS promote what is known as oxidative stress, which is often seen in pathological conditions (283). Whereas estradiol has been shown to have protective effects against oxidative stress in neurons (284), several groups have described the oxidative stress-inducing effects of BPA (62, 211, 215). Here, we demonstrate that BPA alters the gene expression of oxidative-stress related markers Nos2, Sod1, Gpx1 and Hmox1 in a direction indicative of increased oxidative stress. We therefore evaluated whether oxidative stress may be an upstream mediator of BPA actions that leads to the differential responses seen with Npy expression.

The use of anti-oxidant rich NBA media and ROS scavenger TUDCA, likely mediating its effects through JNK inhibition (273), prevented both the increase and the decrease in Npy expression in response to BPA, implicating oxidative stress as an upstream mediator in both forms of Npy dysregulation. However, NDGA, NAC and vitamin B6 only blocked the effect of BPA on Npy upregulation in the mHypoA-59 cells, suggesting the downstream mechanisms activated by oxidative stress in each cell line is different. NAC and vitamin B6 reduce oxidative stress by enhancing anti-oxidant glutathione formation (285, 286). Interestingly, Irrcher et al.

95

(2009) showed that oxidative stress-induced phosphorylation of AMPK in skeletal muscle cells was reverted in the presence of antioxidants, including NAC (287). Thus, it is plausible that

NAC and vitamin B6 may prevent the BPA-induced AMPK phosphorylation in the mHypoA-59 cells, thereby contributing to the prevention of Npy mRNA induction by BPA. Any prevention of AMPK phosphorylation in the mHypoE-46 cells will not block BPA-mediated Npy downregulation as was seen with Compound C (unpublished data, NL and DDB). (see Figure 4-9 for Summary).

BPA BPA mHypoA-59 mHypoE-46

NBA CC NBA NAC Oxidative pAMPK NAC Oxidative TUDCA pAMPK G15 TUDCA Stress B6 Stress SP SP PHTPP NDGA B6 - + - - -

Esr2 Npy Esr1 Esr2 Npy

Figure 4-9 Summary of Npy regulation by BPA in mHypoA-59 and mHypoE-46 cells BPA induces oxidative stress and increases pAMPK levels in both mHypoA-59 and mHypoE-46 cells. Activation of pAMPK is required for BPA-mediated increase in Npy, but not the decrease. Whereas anti-oxidants NBA, TUDCA, SP600125, NDGA, NAC and vitamin B6 prevent BPA- mediated upregulation of Npy expression in the mHypoA-59 cells, only NBA, TUDCA and SP600125 block the BPA-mediated downregulation of Npy in the mHypoE-46 cells. Thus, oxidative stress is implicated in the BPA-mediated dysregulation of Npy, despite the involvement of different downstream events. These differential mechanisms may include activation of GPER or ER in the mHypoE-46 cells as the respective antagonists, G15 and PHTPP, reverse the BPA- mediated decrease in Npy. NBA also blocks the BPA-mediated changes in Esr1 and Esr2 mRNA levels in the respective cell lines, and vitamin B6 mitigates the BPA-induced decrease in Esr2 in the mHypoA-59 cells.

Nevertheless, the finding that supplementation with a vitamin can protect select hypothalamic neurons from BPA-induced Npy dysregulation is noteworthy. Interestingly, one other study described the protective effects of vitamin B6 against BPA-induced glucose intolerance (288), suggesting vitamin B6 supplementation may counteract both hypothalamic and peripheral

96

dysfunction induced by BPA. Whether vitamin B6 can protect against BPA-induced weight gain warrants investigation in animal models and in clinical studies. Indeed, vitamin B6 status was correlated with increased fat-free mass percentage in obese and overweight women on a calorie reduction diet (289), illustrating the potential anorexigenic roles of vitamin B6. Furthermore, direct supplementation with a leucine and vitamin B6 mixture for four weeks decreased plasma oxidative stress markers and increased fat oxidation and insulin sensitivity in overweight and obese men and women (290).

Vitamin B6 and vitamin B12 did not prevent the BPA-induced Npy downregulation in the mHypoE-46 cells, although the NBA media was protective. In addition to the aforementioned anti-oxidant components of NBA, NBA differs from DMEM in that it contains L-alanine, L- asparagine, L-proline, sodium pyruvate and zinc sulfate. L-alanine, sodium pyruvate and zinc have been proposed to have to anti-oxidant properties (291-293), and these may mediate the protective effects of NBA in the mHypoE-46 cells.

As estrogen receptor mRNA dysregulation is also reversed in both cell lines with NBA, this implicates an oxidative stress induced-transcription factor(s) that targets both Npy mRNA and Esr mRNA expression. The expression levels of such transcription factor(s) are likely different between the cell lines, thus contributing to the differential Npy and Esr regulation. The identity of these factors remains to be determined; however, plausible candidates were identified from a previously published gene array analysis of the mHypoE-46 cells compared with the mHypoA- 2/12 cells (294). Although several of these genes were differentially expressed in embryonic versus adult cell lines, heat shock protein beta-1 (Hspb1 or Hsp27) demonstrated an interesting pattern with lower expression in the cell lines that showed a decrease in Npy mRNA in response to BPA (Table 4-1). In addition to its chaperone-like activities, HSPB1 has been described as a transcriptional co-regulator. Specifically, it can interact with the Sp1 transcription factor (295) as well as ER (296), both of which are implicated in Npy transcriptional regulation (95, 297). As ER was implicated in the BPA-mediated Npy downregulation in the mHypoE-46 cells, the higher expression of HSPB1 in the mHypoA-59 cells may prevent BPA-mediated ER activity. These hypotheses warrant further investigation.

One important finding of this study sometimes overlooked in in vitro experiments is that culture media formulations can have a large impact on the response of a cell model to chemicals and

97 hormones. It is well understood that hormones or steroid binding proteins in fetal bovine serum as well as phenol-red can mask the effects of steroid hormone treatments in culture. Thus, phenol-red free media with charcoal-dextran stripped FBS is used in most in vitro studies investigating effects of estradiol and other steroid hormones (298). Here we show that even differences in the vitamin or amino acid composition of culture media can markedly alter the response of a cell. NBA medium, however, did not protect against all BPA-induced changes in gene expression, as Agrp mRNA levels were still upregulated in response to BPA in NBA medium (Appendix C), suggesting only specific pathways are sensitive to the media difference. We emphasize the importance of careful evaluation of culture media components, especially in experiments studying oxidative stress.

There are of course advantages and disadvantages to using cell lines in such studies. While primary cultures contain distinct cell types, allowing for communication between these types, and do not undergo global changes in gene expression due to viral transformation, they are difficult to culture, very few neurons are generated per mouse and cell composition can vary between vehicle and treatment plates (299). Cell lines, representing specific subpopulations of neurons, provide an unlimited supply of cells that allow for detailed mechanistic studies of the direct actions of BPA. These cell lines have been characterized to respond to several hormones and peptides and have been demonstrated to respond in a similar manner to ICV and/or in vivo treatments with these hormones and peptides (299, 300). The question of which subpopulation is a reproductive NPY neuron and which is a feeding-related NPY neuron remains unanswered. From previous experiments on E2, leptin and KISS1 regulation of Npy, the anorexigenic responses of mHypoA-59 (93, 126), mHypoE-42 (95, 278) and mHypoA-2/12 (117) neurons point them towards a feeding-related phenotype due to the ability of estrogen and/or leptin to downregulate Npy expression in these neurons. Considering this, we may hypothesize that the upregulation of Npy in response to BPA in these putative feeding-responsive neurons may contribute to the obesogenic effects of BPA. As for the other lines, a thorough characterization of the normal response to E2, KISS1 and a variety of other hormones may be needed prior to classifying the exact phenotype of the neuron. Nonetheless, the fact that the NPY neurons of the hypothalamus are differentially susceptible to BPA-induced changes in Npy expression is evident regardless of the exact function of these unique subpopulations.

98

In conclusion, we describe that similar to physiological signals, such as leptin, E2 and KISS1, BPA has distinct effects on subpopulations of hypothalamic neurons, again illustrating the heterogeneous nature of the NPY system. Inhibition of oxidative stress protects both subpopulations from BPA-induced Npy dysregulation, although different mechanisms are likely involved downstream. Intriguingly, vitamin B6 was sufficient to mitigate the Npy increase in a subtype of NPY neurons. Although the transcriptional mechanisms remain to be elucidated, this study illustrates the complexity of BPA-induced effects on a hypothalamic cell population controlling both feeding and reproduction and suggests a pathway that can be targeted to alleviate these effects.

99

Chapter 5 BPA alters Bmal1, Per2 and Rev-Erba mRNA and requires Bmal1 to increase Npy expression in hypothalamic neurons

Manuscript published in Endocrinology

Citation: Loganathan N, Salehi A, Chalmers JA, Belsham DD. Bisphenol A Alters Bmal1, Per2, and Rev-Erba mRNA and Requires Bmal1 to Increase Neuropeptide Y Expression in Hypothalamic Neurons. Endocrinology 2019; 160:181-192, doi: 10.1210/en.2018-00881 Contributions: • NL completed experiments and wrote the manuscript. • AS completed experiments in the POMC cell lines (Fig 1 in published manuscript, not included in this thesis). AS also assisted with experiments in Figure 5-4. • JAC generated the immortalized mHypoA-Bmal1-WT and mHypoA-Bmal1-KO cell lines. • DDB provided scientific input, direction and funding. • All authors edited the manuscript.

100

5 BPA alters Bmal1, Per2 and Rev-Erba mRNA and requires Bmal1 to increase Npy expression in hypothalamic neurons 5.1 Abstract

BPA, a ubiquitous, environmental endocrine disruptor, is considered an obesogen. Its role in the hypothalamic control of energy balance, however, remains largely unexplored. As disruption of the circadian clock is tightly associated with metabolic consequences, we explored how BPA affects the components of the molecular circadian clock in the feeding-related neurons of the hypothalamus. In immortalized NPY/AgRP-expressing hypothalamic cell lines and primary culture, we describe that BPA significantly alters mRNA expression of circadian clock genes Bmal1, Per2 and Rev-Erb. Furthermore, we use newly generated Bmal1-knock-out (KO) hypothalamic cell lines to link the BPA-induced neuropeptide dysregulation to the molecular clock. Specifically, BPA increased Npy and Agrp mNRA expression in wildtype hypothalamic cells, whereas the increase in Npy, but not Agrp, was abolished in cell lines lacking BMAL1. In line with this, BPA led to increased BMAL1 binding to the Npy promotor, potentially increasing Npy transcription. In conclusion, we show for the first time that BPA-mediated dysregulation of the circadian molecular clock is linked to the deleterious effects of BPA on neuropeptide expression. Furthermore, we describe novel hypothalamic Bmal1-KO cell lines to study the role of BMAL1 in hypothalamic responses to metabolic, hormonal and environmental factors.

5.2 Introduction

Appropriate control of energy homeostasis is dependent on a properly regulated circadian system (301). The SCN is the body’s central clock system and lesions in this region lead to disruptions in the rhythm of food intake (302). Changes in the timing of feeding has been linked to weight gain in both animal models (154) and epidemiological studies (151, 152). The SCN, entrained by light, is responsible for maintaining the rhythm of clocks in all other regions, including the ARC, an area critical for the regulation of energy balance (148). Factors including a HFD can disrupt the circadian clock in this region, uncoupling it from the SCN (301, 303). Other obesity-promoting factors, such as environmental chemicals (15), acting at the level of the hypothalamus, may also disrupt this synchrony.

101

BPA is an environmental EDC detected in 90-95% of urine samples (228, 232). BPA is considered an obesogen, a class of exogenous factors that dysregulate lipid homeostasis and energy balance, predisposing individuals to weight gain (15). In humans, urinary BPA concentrations are positively correlated to increased BMI scores and comorbidities, such as insulin resistance and type 2 diabetes (45, 47, 228, 229). Several reports describe that prenatal BPA exposure in rodents leads to increased feeding and weight gain in offspring (304). Likewise, adult male and female C57BL/6J mice exposed to BPA for 2 or 4 weeks display increased body weight (62, 305).

The impact of BPA on the brain, especially in the hypothalamus, is only beginning to be explored. Within the ARC, the orexigenic NPY/AgRP and anorexigenic POMC neurons, integrate hormonal and metabolic signals to regulate food intake and energy expenditure (76). The few studies investigating the effect of BPA in the ARC have been with either prenatal or perinatal exposure (57, 183, 184). Indeed, perinatal BPA exposure upregulates the levels of AgRP and downregulates POMC protein levels in neural progenitor cells of newborn mice (184), illustrating the potential orexigenic effects of BPA in the neonatal hypothalamus. We have found that acute exposure of adult and embryonic hypothalamic neurons to BPA upregulates the expression of Npy, Agrp (Chapters 3 and 4) and Pomc (203) mRNA, suggesting an overall dysregulation of the hypothalamic control of energy balance at the level of transcription. Whether this dysregulation is linked to disruption, often tightly associated with metabolic consequences, is unknown.

The expression of the feeding neuropeptides follows a rhythm throughout the day. Pomc gene expression itself exhibits circadian rhythmicity in vitro and in vivo, showing a peak 4 hours after dark phase (ZT 16), and a trough during the day (ZT 4–7) (306). Agrp peaks in vivo in the transition between the light and dark phases, while Npy peaks once in the dark phase and once in the light phase (307). However, in an isolated population of hypothalamic neurons in vitro, only Npy, not Agrp, had significant rhythmicity (167).

These rhythms are controlled by the molecular clock, a series of transcriptional-translational feedback loops involving circadian clock genes. The clock genes Clock and Bmal1 produce the proteins, CLOCK and BMAL1. These transcription factors heterodimerize and bind to E-box promotor elements to upregulate the expression of Per 1–3 and Cry 1–2. Subsequently, PER and CRY proteins form heterodimers that interact with the BMAL1:CLOCK complex, to repress their

102 own transcription. The clock genes cycle in opposite phases, and define daily variations in physiological function, which shapes the circadian rhythm (301). The BMAL1:CLOCK heterodimer also activates the transcription of other clock genes, such as Rev-Erb and , which repress Bmal1 gene expression (308). Approximately 10% of the genome is controlled by these transcription factors, which regulate daily patterns of expression (309, 310). The molecular clock is tightly linked to the control of energy balance as both Clock mutant and Per2 mutant mice develop obesity (160, 161). Furthermore, the saturated fatty acid palmitate, which is elevated in obese states (311), alters circadian clock genes alongside altered hypothalamic neuropeptide expression (137).

Given that circadian dysregulation often underlies metabolic perturbations (301, 312) and that BPA leads to weight gain (62, 305), dysergulates hypothalamic neuropeptide expression and causes metabolic perturbations (313), we hypothesized that BPA alters the expression of clock genes in hypothalamic NPY/AgRP neurons. We also hypothesized that changes in neuropeptide gene expression are dependent on the molecular clock as it is these transcription factors that drive metabolic homeostasis and have led to metabolic consequences when mutated (160, 161). Here, we show that BPA dysregulates gene expression of Bmal1, Per2 and Rev-Erb in immortalized hypothalamic NPY/AgRP-expressing neurons. We also describe the generation and characterization of novel hypothalamic cell lines lacking functional BMAL1 protein, and using these, show that BMAL1 is required for changes in Npy expression. These lines provide tools to examine the role of BMAL1 in a variety of hypothalamic responses to both exogenous and endogenous factors, including chemicals, fatty acids and hormones.

5.3 Results

5.3.1 Circadian clock gene expression is altered in response to BPA exposure

We have previously seen that exposure to the obesogenic endocrine disrupting chemical BPA upregulates Npy and Agrp expression in the mHypoE-41 and mHypoA-59 cell lines. As disruption of the hypothalamic circadian system is associated with changes in energy regulation (166, 303), this prompted us to investigate whether BPA alters the expression of three central clock genes in cell lines where neuropeptide expression is dysregulated. 100 M BPA increased Bmal1 expression from 2 to 8 or 16 hours in mHypoE-41 and mHypoA-59 cell lines, respectively (Fig 5-

103

1). Furthermore, BPA downregulated Per2 expression at 16 and 24 hours in NPY/AgRP- expressing lines. Rev-Erb expression was significantly increased at early time points (1-4 hours) in mHypoE-41 and mHypoA-59 cells, and significantly decreased at later time points in the mHypoA-59 lines (8 hours) (Fig 5-1). Although slight differences exist in the timing and magnitude of response to BPA between each cell line, there is an overall increase in Bmal1, a decrease in Per2 and an upregulation in Rev-Erb, illustrating a dysregulation in the normal expression of these circadian rhythm-associated genes. These effects are replicated in primary culture from CD-1 mice, where 8 hours of BPA treatment upregulated Bmal1 and Rev-Erb in both male and female-derived cultures and downregulated Per2 in female-derived cultures (Fig 5- 1C). Interestingly, despite an initial upregulation in Bmal1 mRNA (Fig 5-1D), Bmal1 shows a trend towards downregulation mHypoE-46 cell line at 16 hours, corresponding to the downregulation in Npy seen in this cell line (Chapter 4, Fig 4-1).

A mHypoE-41

4 Bmal1 Per2 Rev-Erbα

A ****

N Veh

R 3 **** 100 µM BPA

m

7 l

p 2 ****

R ****

/ ****

e n

e 1 G *** *** 0 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24 B mHypoA-59

4 Bmal1 Per2 Rev-Erbα

A ****

N Veh

R 3 100 µM BPA

m

7 l

p 2 R / **** e **** n * ** * e 1 * *** ** G ** 0 2 4 8 16 24 2 4 8 16 24 2 4 8 16 24

104

C Primary Culture (8 h) Vehicle 100 µM BPA

1.5

A ** ** *** **

N R

m 1.0

*

7

l

p

R /

e 0.5

n

e G 0.0 Bmal1 Per2 Reverbα Bmal1 Per2 Reverbα Males Females

D mHypoE-46

2.5 Bmal1 Per2 Veh

** 100 µM BPA

A **

N 2.0

R m

1.5

7

l

p R

/ 1.0

e n e 0.5 * ****

G *** 0.0 2 4 8 16 24 2 4 8 16 24

Figure 5-1 BPA dysregulates circadian gene expression in hypothalamic NPY/AgRP- expressing neurons. (A) NPY/AgRP-expressing mHypoE-41, (B) mHypoA-59 and (D) mHypoE-46 cell lines were treated with 100 µM BPA or vehicle (0.05% EtOH) for 2, 4, 8, 16 and 24 hours and mRNA expression of Bmal1, Per2 and Rev-Erb was measured by qRT-PCR (n=3-4). (C) Hypothalamic primary cultures from CD-1 male and female mice were also treated with 100 µM BPA for 8 hours and mRNA expression measured (n=4). Data are expressed as mean +/- SEM with open circles and bars representing vehicle (0.05% EtOH) and shaded circles and bars representing 100 µM BPA treatment. Statistical significance was determined using a Two-way ANOVA, followed by the Bonferroni post-hoc test (A-C) or a multiple T-test (D); *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001. 5.3.2 Putative binding sites for BMAL1-CLOCK heterodimers exist in the regulatory regions of Npy and Agrp

Considering that BPA-induced alteration of Bmal1 expression precedes, or occurs simultaneously with, changes in Npy or Agrp, we questioned whether BMAL1 may be a mechanistic link between

105

BPA and the changes seen in neuropeptide gene expression. To address this question, we first determined potential BMAL1:CLOCK heterodimer binding sites in the 2500 bp 5’ regulatory regions of neuropeptide genes. BMAL1:CLOCK heterodimer canonically binds to E-boxes 5’CACGTG3’ (314) in DNA; however, it can also bind to other sequences including 5’CACGNG3’, 5’CACGTT3’, 5’CATG(T/C)G3’ (315) and 5’CANNTG3’ (316). Several positions containing the 5’CANNTG3’ sequence are illustrated in Fig 5-2. Binding sites in the Npy and Agrp promotor have been previously reported (167). We identified 4 additional sites in the Npy and 1 additional site in the Agrp promotor. These newly identified sites were located between the -2500 bp and -1500 bp regions of the promotors, except for the 5’CAAATG3’ site at -1217 in the Npy promotor. In total, the regulatory region of Npy contained 10 sites and Agrp contained 16 sites (Fig 5-2). The presence of E-box elements in these regulatory regions suggests the potential action of BMAL1 as a transcriptional regulator of Npy and Agrp.

CAAATG CATGTG A Npy, + strand -1217 -1207 CAGATG CAGTTG CATTTG CAAATG CAGCTG CAAGTG CACCTG Npy CAACTG -433 +106 -2500 -2258 -2058 -1556 -1481 -1125 -694 +1

B Agrp, - strand CACTTG CAAGTG CAGCTG CAGCTG CACATG CAAATG

-1764 -1754 -334 -317 -262 -250 CACATG CACCTG CATTTG CATGTG CAAATG CATCTG CACTTG CATGTG CATCTG CAAATG Agrp

-1951 -1532 -1445-1427 -1130 -811 -526 -462 -399 -112 +1 -2500 Figure 5-2 Potential BMAL1:CLOCK binding sites in the Npy and Agrp promotors. Putative binding sites for BMAL1:CLOCK are represented by gray ovals in the 2500 bp upstream (5’) region of the transcriptional start site (TSS, +1) of the (A) Npy and (B) Agrp genes. The sequence of each binding site is written above the oval with the location relative to the TSS listed below. Where two or more sites are close together, the area is expanded and detailed in the inserts. Lighter ovals represent sites identified previously by Fick et al. (2010) (167) and darker ovals represent newly identified binding sites. All sequences are listed 5’ to 3’.

5.3.3 Characterization of mHypoA-Bmal1-KO cell lines

In order to elucidate the role of BMAL1 on BPA-induced changes in neuropeptide expression, a neuronal model lacking functional BMAL1 protein expression was needed. Having previously

106 encountered difficulty with siRNA knockdown of Bmal1, our laboratory generated immortalized hypothalamic cell lines from female and male whole-body Bmal1-KO mice, alongside WT littermate controls. These cell lines are non-clonal, exhibit neuronal morphology (Fig 5-3B) and express Npy, Agrp and Pomc (Fig 5-3C). The absence of BMAL1 protein and mRNA was verified by Western Blot (Fig 5-3A) and qRT-PCR (Fig 5-3C). Although the knockout models lack Bmal1 expression, basal expression of other genes in the circadian feedback loop (Clock, Per2, Cry, Rev- Erb) were similar to wildtype cells (Fig 5-3C). The expression of inflammatory markers and receptors, including the ERs, ERR훾, and PPAR훾 were verified using qRT-PCR. Besides the lack of estrogen-related receptor gamma (Esrr훾) expression in the mHypoA-Bmal1-KO/M cells, the expression profiles of the cell lines are relatively similar. These receptors have been linked to the mechanism of BPA action in peripheral tissues and neuronal models (203, 270), suggesting the cells are likely to respond to BPA exposure.

107

108

Figure 5-3 Characterization of the mHypoA-Bmal1-WT and -KO cell models. (A) mHypoA-Bmal1-KO/F and mHypoA-Bmal1-KO/M cell lines do not express BMAL1. Expression or absence of BMAL1 protein in the wild-type versus knock-out cell lines was verified with Western Blotting using a BMAL1 antibody and alpha-tubulin as a loading control. (B) Cell lines were imaged using an Olympus CKX41 microscope (10X objective) with the Tucsen 10.0 MP IS1000 USB camera and exhibit neuronal morphology. (C) Summary of circadian, neuropeptide, inflammatory marker and steroid receptor mRNA expression in hypothalamus tissue, mHypoA-Bmal1-WT/F, mHypoA-Bmal1-WT/M, mHypoA-Bmal1-KO/F and mHypoA- Bmal1-KO/M cell lines. RNA was isolated from untreated hypothalamic tissue or cells prior to cDNA synthesis and analysis by qRT-PCR. Relative expression is denoted by + or -, where – (not expressed): cycle at threshold (CT) ≤ 35, + : CT = 30-34.9, ++ : CT = 25-29.9, +++ (highly expressed): CT = 20-24.9; murine (m), hypothalamus (hypo), adult (A), male (M), female (F)

5.3.4 The effect of BPA on Npy expression, but not Agrp, is dependent on Bmal1

Using the mHypoA-Bmal1-KO and wildtype littermate control cell lines, we investigated whether BMAL1 was involved in BPA-mediated upregulation of the neuropeptide gene expression. The mHypoA-Bmal1-KO/F (Fig 5-4A) and mHypoA-Bmal1-KO/M (Fig 5-4B) cell lines were treated with 100 M BPA for 8 hours alongside mHypoA-Bmal1-WT/F controls. Agrp expression was increased at 4 and 8 hours in both wildtype and knockout cell lines (Fig 5- 4). However, the increase in Npy expression seen in the mHypoA-Bmal1-WT/F cell line at 8 hours was not present in neither female mHypoA-Bmal1-KO/F nor male mHypoA-Bmal1-KO/M cell models, implicating the involvement of BMAL1 in the upregulation of Npy, but not Agrp in response to BPA. Interestingly, basal mRNA levels of both Npy and Pomc were altered in mHypoA-Bmal1-KO cells compared to mHypoA-Bmal1-WT/F cells (Fig 5-4C).

109

A mHypoA-Bmal1-KO/F cells B mHypoA-Bmal1-KO/M cells i 4 hours 8 hours i 4 hours 8 hours

**** Legend 2 ** 3

**** mHypoA-BMAL1-WT/F A *** A

N Veh

N R

R 2 ** 100 µM BPA

m

m

7

7 l

1 l mHypoA-BMAL1-KO/F p

p ***

R R

/ or /M

/ y

y 1

p p

N Veh N 100 µM BPA 0 0 WT Bmal1-KO WT Bmal1-KO WT Bmal1-KO WT Bmal1-KO ii Female Female ii Male Male

2.5 2.5

**** A

A *** N

N 2.0 2.0 **** R

R ** *** m

m *

1.5 * 1.5 ****

7

7

l

l

p

p

R

R /

/ 1.0 1.0

p

p

r

r

g g

A 0.5 A 0.5

0.0 0.0 WT Bmal1-KO WT Bmal1-KO WT Bmal1-KO WT Bmal1-KO Female Female Male Male

C i mHypoA-Bmal1-WT/F-Vehicle mHypoA-Bmal1-KO/F -Vehicle 4

Npy Agrp Pomc

A

N 3

R

m

7

l 2

p

R

/

e n

e 1 G

0 4 h 8 h 4 h 8 h 4 h 8 h ii Average Ct Levels for Npy, Agrp, Pomc and Rpl7in Vehicle-treated Samples Cell Line Npy Agrp Pomc Rpl7 mHypoA-Bmal1-WT/F 28.0 28.1 29.8 17.8 mHypoA-Bmal1-KO/F 28.7 27.9 28.1 17.6 mHypoA-Bmal1-WT/F 28.6 27.7 30.1 17.1 mHypoA-Bmal1-KO/M 29.1 27.9 29.1 17.3 Figure 5-4 BPA upregulates Npy and Agrp in mHypoA-Bmal1-WT/F cells, whereas the upregulation in Npy is absent in mHypoA-Bmal1-KO/F and mHypoA-Bmal1-KO/M cells. (A) mHypoA-Bmal1-KO/F (A, n=4-5) and (B) mHypoA-Bmal1-KO/M (B, n=6 for 4 hours, n=3 for 8 hours) cells were treated with 100 µM BPA or vehicle (0.05% EtOH) for 4 or 8 hours alongside mHypoA-Bmal1-WT/F. Npy (Ai and Bi) and Agrp (Aii and Bii) mRNA expression was quantified using qRT-PCR. Data are expressed as mean +/- SEM. Statistical significance was determined using a Two-way ANOVA, followed by the Tukey’s multiple comparison test; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001. (Ci) Npy and Agrp and Pomc expression (n=5) in mHypoA-Bmal1-KO/F cells compared to mHypoA-Bmal1-WT/F cells treated with vehicle (0.05% EtOH) for 4 or 8 hours. All groups are expressed relative to the 4-hour vehicle-treated mHypoA-Bmal1-WT/F group, which was normalized to 1. Data are expressed as mean +/- SEM. (Cii) Mean cycle at threshold (CT) levels of Npy, Agrp, Pomc and Rpl7 as measured by quantitative reverse-transcriptase PCR (qRT-PCR) in vehicle-treated mHypoA-Bmal1-KO/F or mHypoA- Bmal1-KO/M cells compared to mHypoA-Bmal1-WT/F cells that were treated in parallel.

110

5.3.5 BPA treatment increases BMAL1 binding to the Npy promoter

In order to elucidate whether BMAL1 is directly involved in regulating the BPA-mediated induction of Npy by binding to the Npy promoter, we performed ChIP using a BMAL1 antibody in the mHypoA-Bmal1-WT/F cells. Compared to negative control IgG, binding was slightly greater in the chromatin immunoprecipitated with BMAL1, suggesting a basal degree of BMAL1 binding to this region of the Npy promotor at 4 hours. Treatment with 100 M BPA significantly increased the amount of BMAL1 protein binding to the -1267 to -1165 region of the Npy promotor compared to vehicle treatment after 4 hours (Fig 5-5B). Two putative binding sites exist for the BMAL1:CLOCK heterodimer in this region of the promotor, including 5’CAAATG3’ at -1217 and 5’CATGTG3’ at -1207 (Fig 5-5A, 5-2A). This suggests that BMAL1 may be involved in regulating Npy at the transcriptional level upon BPA exposure.

A F primer NPY

-1267 -1217 -1207 -1165 +1 R primer

B

2

5 *

6

1

g

1

n

-

i

d

o

t

n

i

7

b

6 1

e

2

v

i

1

t

-

a

l

y

e

p

R

N

t a 0 Veh 100 µM IgG BPA BMAL1

Figure 5-5 BPA increases BMAL1 binding to the Npy promotor. (A) Schematic representing the 103 base pair region of the Npy promotor, containing two potential BMAL1 binding sites (gray ovals), amplified for the ChIP assay using qRT-PCR, F= forward, R= reverse. (B) mHypoA-Bmal1-WT/F cells were treated with 100 µM BPA or vehicle (0.05% EtOH) for 4 hours, and changes in BMAL1 binding to the -1267 to -1165 region of the Npy promotor were measured with a chromatin immunoprecipitation (ChIP) assay (n=3). A vehicle treated chromatin sample was incubated with normal IgG antibody as a negative control. Data is expressed as mean +/- SEM. Statistical significance was determined using a One-way ANOVA, followed by the Bonferroni multiple comparison test; *P<0.05.

111

5.4 Discussion

Under normal conditions, the SCN synchronizes the feeding centres of the hypothalamus, allowing neurons within these regions to respond appropriately to timed signals from the rest of the body (148, 301). However, metabolic factors may dysregulate the expression of clock genes, leading to a loss of synchrony (301). Such factors include high glucose and dietary fatty acids, which can alter the expression patterns of Bmal1, Per2 and Rev-Erb by changing the peak amplitude and period of the 24-hour expression profile (166, 317). In turn, this alters expression profiles of genes that are rhythmically controlled by these clock-associated transcription factors (166, 309, 310). Here, we describe changes in hypothalamic circadian gene expression induced by BPA, a chemical known to lead to metabolic perturbations. In immortalized hypothalamic cell lines and primary culture, we show that exposure to BPA increased Bmal1 expression within 2 to 4 hours and decreased Per2 expression at 16 and 24 hours. This was observed across several hypothalamic cell lines, representing both adult and embryonic cells as well as orexigenic NPY/AgRP-expressing cells and anorexigenic POMC-expressing cells (data not shown here, (225)), illustrating the robust nature of these effects across heterogeneous populations of neurons. Rev-Erb was also altered, although the pattern was more heterogeneous across the cell lines.

Disruptions in daily patterns of activity have been reported with gestational BPA exposure in zebrafish and mice (318, 319). Although there are no studies to date investigating circadian gene dysregulation by BPA in the adult mammalian hypothalamus, Sellix et al. reported altered clock gene expression in reproductive tract tissues (320). Furthermore, Choi et al. showed that Per2 and Cry1 levels were downregulated with increasing BPA concentration and exposure time in goldfish brain and liver (321). Circulating concentrations of melatonin, a hormone involved in entrainment of circadian rhythms, as well as the expression of the (MT-R1) were also decreased with BPA exposure (321). The impact of endocrine disrupting chemicals besides BPA on circadian systems, particularly pertaining to reproductive tissues, has been described (320). Interestingly, prenatal exposure of Sprague-Dawley rats to a weakly estrogenic polychlorinated biphenyl mixture (Aroclor 1221) altered Bmal1 and Per2 expression in the AVPV of the hypothalamus in female pups (322). These studies, along with our work, illustrate that the circadian clock system is vulnerable to dysregulation by endocrine disrupting chemicals, including BPA.

112

A dysregulation in clock gene expression has broad implications for cellular function, such as cell migration (323) and gene transcription (309). Changes in core clock genes by BPA may disrupt the transcription-translation feedback loop that exists within these hypothalamic neurons, thereby disrupting rhythmicity. Circadian clock genes not only act to maintain a 24- hour rhythmicity within the body, they bind to promotor regions of up to 10% of genes and can alter expression of these several E-box domain containing genes (309, 310). We and others describe the presence of potential BMAL1:CLOCK heterodimer binding sites on the Npy and Agrp (167) gene promotors, suggesting that BPA-induced disruption of components of the molecular clock may be responsible for the changes observed in Npy and Agrp expression.

To experimentally determine the role of clock genes in the regulation of feeding neuropeptides by BPA, we chose a Bmal1-KO model. BMAL1 is critical for the circadian regulation of energy homeostasis as whole-body Bmal1-KO mice exhibit disruptions in the daily rhythm of glucose and triglyceride levels (324). Unlike Cry and Per knock-out mice, Bmal1-KO mice lack functional redundancy since BMAL1 controls the transcription of the paralogous gene Bmal2, leading to the complete absence of BMAL function (325, 326). Using previously established techniques (220), we immortalized hypothalamic neurons from Bmal1-KO animals and their WT control littermates. With the generation of these cell models, we showed that BPA-mediated upregulation of Npy, but not Agrp, requires functional BMAL protein expression. This may imply that the inability of BPA to change Bmal1 expression, synthesis and/or binding protects the mHypoA-Bmal1-KO hypothalamic cells from BPA-induced dysregulation of Npy expression. While Agrp contains several putative BMAL1 binding sites and may be basally regulated by rhythmic BMAL1 binding, the BMAL1-dependency of the BPA-induced changes is specific to Npy. However, the fact that the Agrp effect is not altered in mHypoA-Bmal1-KO lines coincides with the finding that Agrp mRNA was previously found to lack rhythmicity in one population of hypothalamic neurons (167). The role of BMAL1 in basal Npy, but potentially not Agrp, expression is further illustrated in the differences in basal Npy mRNA levels in the mHypoA-Bmal1-KO cell lines compared to the mHypoA-Bmal1-WT/F cell line (Fig 5-4C). As BMAL1 rhythmically binds to the Npy promotor (167), the lack of BMAL1 expression in the mHypoA-Bmal1-KO cells may explain the slightly lower expression of Npy. Interestingly, the Pomc promoter also contains putatative BMAL1 binding sites (data not shown here). The difference in basal expression of Pomc between the mHypoA-Bmal1-WT/F and the mHypoA-Bmal1-KO cell lines does not alter the magnitude of

113 the BPA-induced Pomc response, again emphasizing the specificity of the BMAL1-dependent effect of BPA on Npy.

Others have alluded to the relationship between Bmal1 and Npy. For example, a low protein diet in pregnant dams ablated the circadian rhythm of Bmal1 expression in the hypothalamus of 17- day old offspring, with a concurrent shift in the diurnal rhythm of Npy expression (327). Furthermore, in a clonal Npy-expressing cell line, palmitate increased Bmal1 expression and decreased its amplitude of oscillation along with an increase in Npy mRNA levels (166). However, in these studies, it remained unclear whether BMAL1 has direct transcriptional control on Npy expression. Fick et al. demonstrated a rhythmic pattern of BMAL1 binding to the Npy promotor across a 24-hour time course in the mHypoE-44 hypothalamic cell line, suggesting this rhythmic binding may contribute to the rhythmic expression of Npy mRNA (167). Specifically, using ChIP analysis, we have shown here that BPA increases the relative amount of BMAL1 binding to the Npy promotor prior to the changes seen in Npy expression, suggesting transcriptional upregulation of Npy. In line with our findings, nervous system-specific Bmal1-KO mice have decreased body weight and food intake (328). As increased BMAL1 binding may contribute to increased expression of Npy, downregulation of Npy may partially underlie this anorexigenic phenotype of Bmal1-KO mice.

Whether factors other than BMAL1 lead to the lack of BPA-induced Npy expression in the mHypoA-Bmal1-KO cells warrants investigation. The mHypoA-Bmal1-KO and mHypoA- Bmal1-WT cell lines are similar in terms of basal expression of neuropeptides, circadian rhythm genes, and inflammatory genes. The lack of Esrr훾 expression in the mHypoA-Bmal1-KO/F cells likely does not contribute to the ablated Npy response since the mHypoA-Bmal1-KO/M cells express Esrr훾 at the same level as mHypoA-Bmal1-WT cells. Although basal expression levels of these genes are similar across both knockout and wildtype cell lines, their rhythmic expression over 24 hours may be affected by the absence of BMAL1 (168). For instance, rhythmic expression of circadian genes Rev-Erb and D-box binding PAR bZIP transcription factor (Dbp) are abolished in adrenal cortex Bmal1-KO animals (329). Future investigations may include whether changes in rhythmic expression profiles of certain genes contribute to the abolished Npy response. Regardless, ablation of a crucial circadian gene is not a feasible or desirable method to protect hypothalamic cells from BPA-mediated neuropeptide dysregulation. Elucidating the mechanism

114 by which this dysregulation in circadian gene expression and/or binding occurs may provide beneficial information to block the effects.

The mechanisms by which BPA alters clock gene expression in feeding-related hypothalamic neurons remains unknown. BPA has primarily been thought to act through nuclear and membrane- bound ERs, and other nuclear receptors, including PPAR훾, estrogen-related receptor gamma, AhR, AR and GR (270, 330). These receptors can bind to response elements in promotors and act as transcription factors to regulate gene expression. In fact, Rhee et al. described the presence of xenobiotic response elements, including AhR response elements and estrogen receptor response elements, in the promotor regions of killfish Clock, Bmal1, Per2, Cry1 and Cry2 (331). Furthermore, PPAR훾 has been shown to act as a transcriptional regulator of Bmal1. We have recently identified PPAR훾 as a mediator of BPA-induced upregulation of Pomc in the mHypoA- POMC/GFP-2 cell line (203). In addition to steroid and nuclear receptor activation, BPA can induce inflammatory signaling (203, 305, 332), oxidative stress (62), EndR stress, and MAPK signaling (189, 305, 332). Whether these receptors and pathways are involved in BPA-mediated Bmal1, Per2 and/or Rev-Erb dysregulation in hypothalamic feeding-related neurons remains to be determined.

In conclusion, this study is the first to explore the direct effects of BPA on circadian clock gene expression in NPY/AgRP mammalian hypothalamic cell models, demonstrating altered expression of circadian rhythm genes. We also describe the development, characterization and experimental use of novel Bmal1-KO hypothalamic cell lines that can be used to define the role of BMAL1 in hypothalamic responses to various hormonal and metabolic signals. Using these cell lines, we show the involvement of BMAL1 in BPA-mediated Npy dysregulation, implicating hypothalamic circadian rhythm alterations in the metabolic consequences of BPA exposure. These findings in turn shed light on the fundamental connection between metabolism and the circadian molecular clock.

115

Chapter 6 Discussion

116

6 Discussion 6.1 Summary of Findings

BPA has been well established as a disruptor of reproductive function at the level of peripheral cells and at the level of the hypothalamus. It has more recently been described as an obesogen, with limited evidence suggesting perinatal exposure leads to an upregulation in Npy and Agrp (57, 184). Furthermore, Desai et al. reported an increase in Npy and Agrp protein levels when neural progenitor cells were treated with BPA in vitro (184). No studies had investigated the effects of BPA on the adult NPY/AgRP neuronal system and moreover, the existing studies lacked the mechanistic basis underlying the BPA-mediated dysregulation of Npy and Agrp. The specific mechanisms underlying the dysregulation of Agrp and Npy have been described in the previous chapters; thus, the following section will focus on the links between these pathways and the subsequent consequences of such dysregulation.

To first summarize the findings, BPA increased Agrp expression in all cell lines studied, while BPA increased Npy expression in five cell lines (mHypoA-59, mHypoA-2/12, mHypoE-41, mHypoE-42, mHypoA-BMAL1-WT/F), decreased Npy expression in two cell lines (mHypoE- 46, mHypoE-44) and did not alter Npy expression in two cell lines (mHypoA-BMAL1-KO/F, mHypoA-BMAL1-KO/M). The upregulation in Agrp required pAMPK and the transcription factor ATF3, while the upregulation in Npy involved oxidative stress, pAMPK and the transcription factor BMAL1. Finally, Npy expression was downregulated by BPA in specific Npy-expressing cells via GPER and ER (Fig 6-1). Using the clonal cell models, any differences in response between embryonic- vs. adult-derived or male- vs. female-derived cell lines can be ascertained (Table 6-1). Notably, although the fold change in Agrp expression was similar in all cell lines tested, embryonic-derived cell lines responded earlier and at more time-points than adult-derived cell lines, which demonstrated maximal induction in Agrp at 8 hours. In terms of Npy, the magnitude of induction was greater in adult-derived cell lines compared with embryonic-derived cell lines. There were no apparent differences between female- and male- derived lines, besides the fact that only embryonic, male-derived cell lines had decreased Npy levels after BPA exposure. It is difficult to conclude whether this is specific to male-derived lines due to the limited availability of embryonic, female-derived cell lines expressing Npy. Overall, these results suggest that embryonic-derived lines may be more susceptible to BPA-induced

117 changes in Agrp, while adult-derived cell lines respond more strongly with BPA-mediated Npy induction. This may imply differences in the orexigenic effects of embryonic vs. adulthood exposure to BPA based on the specific function of each neuropeptide.

Table 6-1 Comparison of BPA-induced effects on Npy and Agrp expression in (A) embryonic- vs. adult-derived and (B) female- vs. male-derived cell lines

A

Embryonic- vs. Adult-Derived Cell Lines

Gene Embryonic-derived: Fold change (time) Adult-derived: Fold change (time) Agrp mHypoE-41: ↑ 1.7x (2 h) mHypoA-59: ↑ 1.6x (8 h) mHypoE-42: ↑ 1.7x (2, 8, 16 h) mHypoA-2/12: ↑ 1.3x ( 8 h) mHypoE-46: ↑ 1.4x (2, 8 h) mHypoE-44: ↑ 1.7x (2-8 h)

Comparison BPA increased Agrp at an earlier time point in embryonic-derived cell lines

Npy mHypoE-41: ↑ 2.5x (8, 16 h) mHypoA-59: ↑ 9x (16 h) mHypoE-42: ↑ 1.8x (8, 16 h) mHypoA-2/12: ↑ 4x (8 h) mHypoE-46: ↓ 2x or 50% (16 h) mHypoE-44: ↓ 7x or 86% (24 h)

Comparison Magnitude of BPA-mediated increase in Npy was larger in adult-derived cells

B

Female- vs. Male- Derived Cell Lines

Gene Female-derived: Fold change (time) Male-derived: Fold change (time) Agrp mHypoE-41: ↑ 1.7x (2 h) mHypoE-42: ↑ 1.7x (2, 8, 16 h) mHypoA-59: ↑ 1.6x (8 h) mHypoE-46: ↑ 1.4x (2, 8 h) mHypoE-44: ↑ 1.7x (2-8 h) mHypoA-2/12: ↑ 1.3x ( 8 h) No major differences between male- and female-derived cell lines in BPA-mediated Comparison Agrp induction Npy mHypoE-41: ↑ 2.5x (8, 16 h) mHypoE-42: ↑ 1.8x (8, 16 h) mHypoA-59: ↑ 9x (16 h) mHypoE-46: ↓ 2x or 50% (16 h) mHypoE-44: ↓ 7x or 86% (24 h) mHypoA-2/12: ↑ 4x (8 h)

Comparison BPA-mediated downregulation in Npy occured only in male-derived cells

Fold or % change listed is the maximal fold or % change that was observed over a 24 h time course. Time point(s) at which the maximal effect occurred is indicated in brackets.

118

6.2 Overall Discussion

Regardless of the different transcription factors involved in Agrp versus Npy regulation and the different mechanisms mediating the responses between the three Npy populations studied (see Appendix A, C, D for mHypoE-41 and Fig 6-1), compound C prevented the upregulation of Npy and Agrp in both mHypoA-59 and mHypoE-41 cells, implicating AMPK as an upstream mediator of BPA-induced Agrp and Npy and upregulation. While AMPK was activated in mHypoE-46 cells, compound C did not prevent the BPA-mediated downregulation of Npy.

The activation and downstream activity of AMPK largely depends on the tissue type examined. This is primarily due to the different subunits of AMPK expressed and upstream kinases active in each tissue type. AMPK is a heterotrimeric protein, composed of an α-catalytic subunit (α1, α2), a ß-regulatory subunit (ß1, ß2) and a -regulatory subunit (1, 2, 3). α2 is highly expressed in neurons compared with α1, ß2 is predominantly expressed in skeletal muscle, 3 expression is restricted to skeletal muscle, and 1 is dominant in neurons (333, 334). Importantly, the expression of these subunits is altered throughout development in the mouse brain: α2 and ß2 expression increases between embryonic day 10 and 14, while levels of α1, ß1 and 1 remain consistent through this time into the postnatal period (335). The type of -subunit present determines affinity of AMPK to AMP binding, which allosterically activates AMPK. Upon binding, a conformational change occurs exposing Thr172 on the α-catalytic subunit, enabling phosphorylation by upstream kinases, liver kinase B1 (LKB1), CAMMKß or transforming growth factor-beta activated kinase 1 (TAK1), and subsequent 50- to 100-fold increased AMPK activity (333, 334). LKB1 primarily acts as an upstream kinase in liver and muscle cells (333), while CAMKKß was shown to mediate Ca2+-dependent phosphorylation of AMPK in rat brain slices (333, 336), which represents a mechanism through which oxidative stress can activate AMPK. The functional differences in subunits is also evident in global KO studies. While global KO of the ß1 subunit decreased food intake and body mass (337), global KO of the ß2 subunit, primarily expressed in the muscle (338), led to increased weight, increased insulin levels and glucose intolerance (339), illustrating the distinct actions of different AMPK isoforms. Regardless, AMPK is activated by signals that deplete ATP levels to ultimately generate ATP. At the level of the hypothalamus, this need to increase ATP levels manifests with increased food intake, and several orexigenic signals activate AMPK (334).

119

AMPK has been shown to modulate both ATF3 (as described in chapter 3) and BMAL1. The AMPK-mediated increase in ATF3 activity likely leads to the induction of Agrp in both mHypoA-59 and mHypoE-41 cells and the induction of Npy in the mHypoE-41 cells as evidenced by the ability of compound C to inhibit Atf3 upregulation by BPA. While the upregulation in Npy in the mHypoA-59 cells or the mHypoE-41 cells cannot be conclusively linked to BMAL1 since BMAL1 was not knocked out of these cell lines, the existing relationship between AMPK and BMAL1 alludes to the strong possibility (Fig 6-1). As an energy sensing molecule, AMPK phosphorylates negative regulators of the CLOCK:BMAL1 heterodimer – CRY and CKIε (which degrades PER2), leading to stabilization of CLOCK:BMAL1 (340). Furthermore, AMPK activation using AICAR was shown to lead to a phase shift in BMAL1 expression in fibroblasts (341) and mice lacking BMAL1 had reduced AMPK-mediated glucose uptake in skeletal muscle (342). These experiments verify the interaction between AMPK and BMAL1 and point to the possibility that AMPK activation may lie upstream of increased BMAL1 binding to the Npy promoter seen in Chapter 5. If BPA activates AMPK in these hypothalamic neurons, the fact that the lack of BMAL1 prevented BPA-mediated changes in Npy expression is reminiscent of the above-mentioned study of reduced AMPK-mediated glucose uptake and in turn, points to the involvement of BMAL1 in AMPK-mediated Npy upregulation in the mHypoA-59 and mHypoE-41 cells. Interestingly, with compound C, basal levels of Bmal1 were significantly downregulated in mHypoA-59 and mHypoE-41 cells (unpublished, NL and DDB). Whether BMAL1 binds to the promoter of Npy in mHypoA-59 neurons and whether AMPK inhibition prevents this effect is an avenue for future investigation. In the mHypoE-41 cells, AMPK is likely activated by another upstream pathway, independent of oxidative stress, as blocking oxidative stress does not mitigate BPA-induced Npy upregulation in these cells (Appendix D). Increased intracellular calcium as a result of the interaction of BPA with plasma membrane and intracellular calcium ion channels (215, 343, 344), followed by CAMMKß activation and AMPK phosphorylation is one such alternative pathway that warrants investigation.

It is intriguing that in the mHypoE-46 cells, Bmal1 expression shows a trend towards downregulation at 16 hours, the same time point at which BPA leads to Npy downregulation. Thus, it is possible that reduced BMAL1 binding to the Npy promoter in these cells prevents BPA-mediated Npy upregulation. However, it was described in Chapter 4 that Npy

120 downregulation required ERß and GPER. It has been reported that Bmal1-KO mice lose rhythmic expression of ERß in skeletal muscle, but have greater basal ERß expression compared to wildtype mice (345). Contrary to this finding, mHypoA-BMAL1-KO/F and /M cells had decreased basal ERß compared to wildtype hypothalamic neurons (Appendix H). The fact that BPA did not lead to an upregulation or a downregulation in Npy in the mHypoA-BMAL1-KO/F and /M cells may be a result of the interplay between the lack of BMAL1 required to upregulate Npy and the decreased or dysregulated ERß which may be required to downregulate Npy.

The complexity of the mechanisms activated by BPA in these hypothalamic neurons makes therapeutically targeting a single pathway to block all effects impossible. However, the orexigenic effects seem to be dependent on AMPK and the Npy downregulation is likely mediated by estrogen receptor activation, pointing to possible avenues to block specific effects. Acting as potent orexigens, an increase in Agrp and Npy expression can lead to obesity as a result of increased food intake and decreased energy expenditure. However, both Agrp and Npy can alter energy homeostasis independent of food intake (80). For instance, pair feeding to prevent differences in food intake in mice given exogenous NPY still led to metabolic and hormonal abnormalities, including elevated leptin levels (346). Central NPY administration has been shown to promote fat deposition by increasing the activity of acetyl carboxylase in white adipose tissue and triglyceride and fatty acid synthesis in the liver (80, 81, 89, 91). NPY administration was also accompanied by increased circulating insulin levels (347). Likewise, AgRP peptides lacking the C-terminus domain, which binds to MC4, did not affect food intake, but led to decreased energy expenditure, increased body weight and increased epididymal fat in rats (348). Optogenetic activation of AgRP neurons led to insulin resistance in mice via prevention of insulin-stimulated glucose uptake in brown adipose tissue (88). Indeed, BPA has been shown to increase adipogenesis and fat mass (44, 60, 65), lead to insulin resistance and is linked to higher circulating insulin levels (47, 66, 263). As such, the effects of BPA to increase Npy and Agrp expression may not only underlie the BPA-induced increase in body weight due to hyperphagia, but also BPA-induced metabolic abnormalities, such as lipid accumulation and insulin resistance.

The NPY/AgRP neuron is an important target of insulin and leptin to relay satiety to the hypothalamus. NPY/AgRP neurons respond by decreasing both Npy and Agrp expression in response to these hormones, leading to decreased orexigenic signaling and increased anorexigenic MC4 signaling (80). Ultimately, this results in decreased food intake and increased

121 energy expenditure (80). The increase in Agrp and Npy by BPA in a cell line that has previously been shown to respond to leptin by downregulating Npy secretion (mHypoA-59) (126) indicates the potential of BPA to block leptin-mediated Npy and Agrp downregulation. An example of this phenomena was seen in NPY neurons from the amygdala that control food intake. Whereas insulin repressed Npy in this population, in the presence of Npy-increasing factors (high calorie- diet and stress), insulin was unable to elicit its normal repressive effects (349). Such effects can occur as a result of receptor downregulation or activation of transcription factors by the exogenous factors (i.e BPA, high calorie-diet, stress) that maintain high levels of Npy or Agrp expression even in the presence of insulin and/or leptin. It is plausible that aberrant activation of AMPK and subsequent ATF3/BMAL1 activation by BPA can override the effects of the satiety hormones.

In contrast, ghrelin, secreted from the stomach, relays hunger signals to the hypothalamus and is a potent orexigen (350). The effects of BPA on circulating ghrelin levels remains controversial as both positive (351) and negative correlation (352) has been reported. NPY/AgRP neurons are primary mediators of the orexigenic effects of ghrelin as antibodies or antagonists against NPY and AgRP blocked ghrelin-induced food intake (353). NPY/AgRP neurons, including the cell models used in this thesis, express the ghrelin receptor, GHSR. It was reported that ghrelin- mediated increases in Npy and Agrp expression only occurs in the presence of glucocorticoids (350), which is noteworthy as BPA is associated with altered cortisol levels (354, 355). Additionally, ghrelin activates pAMPK (356) and glucose-mediated suppression of ghrelin- induced gene expression in AgRP neurons requires AMPK (357), illustrating the possible convergence of ghrelin- and BPA-mediated signaling. Given that ghrelin is a potent orexigen, the effects of BPA on GHSR levels and any additive effects of ghrelin and BPA on hypothalamic neurons are important topics for investigation.

It is noteworthy that Agrp was still upregulated by BPA in the two cell lines that showed a downregulation in Npy expression. Npy expression is upregulated by the fatty acid palmitate in both the mHypoE-46 and mHypoE-44 cells (138), and leptin and insulin downregulate Npy secretion and expression, respectively, in the mHypoE-46 cells (123, 131). This suggests that these two cell populations also respond to feeding signals, and may be representative of NPY subpopulations that also control feeding - although reproduction and feeding are intrinsically linked, and altered energy balance will alter reproductive function. Ruud et al. recently described

122 that the rapid induction of food intake and the development of insulin resistance that occurs subsequent to NPY/AgRP neuron activation using chemogenetic or optogenetic methods were lost in mice lacking NPY (358). However, a delayed increase in feeding was still present and when NPY was re-expressed, the mice were no longer protected from insulin resistance (358). Thus, NPY may be responsible for immediate feeding in response to a stimulus, whereas AgRP may elicit a more prolonged increase in food intake. This hypothesis is also supported by a previous study where a single intracerebroventricular injection of AgRP was able to increase food intake for up to 7 days (86). As such, it is plausible that the decrease in Npy and the increase in Agrp in the mHypoE-46 cells and mHypoE-44 cells may result in an initial rapid decrease in food intake, with an increase in delayed food intake. We have also reported that in hypothalamic Pomc-expressing cell lines, BPA acutely increases the expression of Pomc at 4 hours. This may represent an acute anorexigenic effect as no further upregulation in Pomc is seen at later time points (203). However, a more sustained, orexigenic effect is suggested as Npy and Agrp were increased at later time points, and other in vitro and in vivo models have demonstrated BPA- induced suppression of POMC with longer treatment times (57, 183, 184).

Stoker et al. recently showed that perinatal exposure to BPA increased food intake, weight gain and fat mass in male rats, accompanied by a paradoxical downregulation in Npy expression. These animals were exposed to BPA from day 9 of gestation to weaning, and Npy expression was measured at 20 weeks of age (359). In contrast, McKay et al. described an increase in both Npy and Agrp expression in adult male mice perinatally exposed to BPA and subsequently fed a HFD compared to those given HFD alone (57). Thus, while these changes in Npy expression in adulthood may result from the metabolic dysfunction caused by perinatal BPA rather than the direct action of BPA on Npy itself, it serves as an example of the complexity of how the NPY system may be differentially altered despite increased body weight and/or disrupted glucose tolerance. Whether insulin signaling is differentially affected in mHypoE-46 and mHypoE-44 neurons compared to cells showing BPA-mediated upregulation in both Npy and Agrp necessitates investigation.

Ultimately, the functions of these subtypes need to be determined by double-labeled staining of NPY and downstream afferent neurons (i.e. a reproductive NPY neuron would project to GnRH neurons) in order to determine the directionality of the Npy response to BPA in reproductive versus feeding-related NPY neurons. Alternatively, hypothalamic slices from NPY/GFP mice

123 can be generated and GnRH can be labelled to visualize the changes in NPY neurons projecting to GnRH neurons. The in vitro studies presented in this thesis report for the first time the existence of such divergent effects, regardless of the specific functions, and shed light on the mechanisms underlying the differences.

124

125

Figure 6-1 Summary of Findings (A) In the mHypoA-59 cells, BPA increases Agrp mRNA levels through AMPK and ATF3 activation, whereas oxidative stress-induced AMPK activation and subsequent BMAL1 promoter binding likely mediate the induction of Npy. (B) In the mHypoE-41 cells, BPA increases Agrp expression again through AMPK and ATF3 activation, however, the induction of Npy is not dependent on oxidative stress. ATF3 mediates the induction of Npy in the mHypoE-41 neurons, and BMAL1 binding to the promoter of Npy is likely involved. It is currently unclear whether ATF3 directly binds to the Agrp and Npy promoters or alters a secondary transcription factor that binds to the promoter region. (C) In the mHypoE-46 cells, the downregulation of Npy by BPA requires estrogen receptors ERß and GPER. Oxidative stress induction and decreased BMAL1 binding to the Npy promoter region may also be involved in this BPA-mediated downregulation of Npy. The presence or absence of transcriptional co-regulators in the different cell lines likely influence how each subpopulation responds to BPA.

6.3 Limitations

Immortalized neurons allow for mechanistic studies of the effects of BPA on Npy and Agrp expression; however, the use of these cell lines comes with limitations that need to be considered when making conclusions. Firstly, in vitro studies using clonal cell lines lack afferent and efferent synaptic connections between neuronal cells as well as the multitude of glial cells characteristic of an intact hypothalamus. Thus, the conclusions derived from this thesis only apply to how the individual hypothalamic neuron responds to BPA and the internal, cellular mechanisms underlying these responses. In an in vivo setting, these responses may be also modulated by the presence of glial cells (360) and circulating hormones, including, but not limited to insulin, E2 and leptin. Secondly, SV-40 large T-antigen was used to immortalize the cell lines. SV-40 T-antigen sequesters and inhibits tumor suppressors p53 and pRB, promoting cell cycle progression into the S-phase and cell proliferation. P53 and pRB are both transcription factors, and along with adaptor proteins that T-antigen has been shown to interact with, p300 and CBP, they have the ability to alter the transcriptional profile of a cell. Furthermore, T-antigen can alter the histone acetylation and DNA methylation status of genes, leading to altered chromatin structure and increased transcriptional capacity (as reviewed in (299)). Using fibroblasts and erythrocytes, Cantalupo et al. have reported that up to 9% of genes analyzed using a microarray were dysregulated with SV-40 T-antigen transformation (361). Interestingly, several genes were regulated in a cell-type specific manner (361). The random nature of SV-40 T-antigen integration into the host genome leads to this cell-specific genetic manipulation that needs to be determined for each experimental model. One such genetic manipulation by SV-40 T-antigen in our cell

126 models is the induction of Agrp expression, of which high levels are found in almost all of our models. siRNA knockdown of T-antigen was able to reduce these levels (unpublished, DDB) (299). Regardless, Agrp expression has been shown in our cell lines to be regulated by hormones, such as glucose, insulin and E2 (95, 131, 139). To circumvent some limitations, cell lines that co-express Npy with Agrp, and thus are more representative of Agrp-expressing neurons in vivo, were used in these studies. Furthermore, appropriate use of vehicle controls allowed for changes in Agrp as well as all other genes and proteins tested to be attributed to the effects of BPA rather than the effects of SV-40 T-antigen. The consistent BPA-induced upregulation in Agrp across cell models, in primary culture and in vivo studies strengthens the results derived from the immortalized cell lines.

While culturing non-transformed primary hypothalamic neurons would circumvent any T- antigen- mediated effects, there are many limitations to primary hypothalamic cultures as well. The mere amount of hypothalamic tissue needed to generate cultures to perform detailed mechanistic studies renders this method impractical. Each hypothalamus can generate up to two wells (6-well dishes) of neurons that typically fill the well within 7 to 9 days. These cells cannot be passaged and propagated without losing subpopulations of neurons, and each well often contains a distinct array of neuronal populations, sometimes making vehicle-treated wells different in cellular composition from BPA-treated wells (299). Thus, we only performed a limited primary culture study where the effects of one concentration of BPA on Agrp, Npy and the circadian genes were determined at one time point. Whereas Agrp and the circadian genes were generally regulated in a similar manner by BPA in primary culture as in the cell models, Npy expression was not altered with 100 µM BPA at 8 hours in primary culture (see Appendix E). The great advantage of having different types of neurons represented is also a caveat of primary culture as effects in individual neuronal subpopulations can be cancelled out by opposite effects in other subpopulations. While the response of each subpopulation would have distinct consequences in an in vivo environment where each neuron would have efferent connections to higher order neurons in the feeding or reproductive circuitry, in in vitro settings where the mRNA of all the neurons are pooled, these changes in cell-specific mRNA levels can be masked. As such, it is not surprising that Npy expression, which is changed differentially in specific cell lines, did not show altered expression in primary culture. In situ hybridization to detect and

127 localize whether there are differential intensities of Npy expression between cell populations after BPA treatment in primary culture will enhance the conclusions derived from this thesis.

Due to the aqueous nature of cell culture media, these studies were also limited by the concentration of BPA that needed to be used. As described in chapter 3, only 18% of BPA was found in the cellular portion with 100 µM treatment for 24 hours, suggesting lower entry of BPA at 8 and 16 hours, where the effects on Npy and Agrp occurred. Most striking was that this percentage of entry was decreased at lower concentrations of 10 µM (14% entry at 24 hours and 2% entry at 4 hours). The amount of BPA measured in the cellular portion represents any BPA adsorbed or bound to the membrane as well as any BPA within the cytoplasm. Thus, the question of how much BPA actually entered into the cytoplasm of the cell remains unanswered. In a zebrafish embryo model, treatment with 100 to 500 µM BPA led to 10 to 30% of BPA adsorbed to the cell membrane. Approximately 10 to 20% of the adsorbed BPA entered the cytoplasm, indicating that BPA can cross cellular membranes in vitro (362). Transmembrane transport of BPA was also observed with radiolabeled BPA in a trophoblast cell line (363).

It was assumed that the majority of the effects elicited by BPA on Npy and Agrp occurred intracellularly in these hypothalamic neurons as evidenced by the elucidated mechanisms. As such, micromolar concentrations were used to ensure adequate entry of BPA into cells. Since different effects were observed between cell lines and distinct mechanisms were identified for each neuropeptide, it is not likely that the higher concentrations of BPA used in these studies led to broad non-specific effects. A similar study using hypothalamic neuroprogenitor cells showed increased cell proliferation as well as increased NPY and AgRP protein levels using 10 and 20 µM BPA treatment over 5 days. The increase in cell proliferation, which was also seen in vivo, was lost at 1 µM in vitro (184), again pointing to the need to use higher concentrations.

Human exposure levels to BPA generally correlated to serum or urine concentrations suggest nanomolar and low micromolar concentrations (up to 3 µM in urine) (36). These do not necessarily account for accumulation of lipophilic BPA in tissues, including the brain (37, 39). It has been reported that BPA can displace labelled estradiol from the plasma sex steroid binding protein (SBP) of rainbow trout (18), suggesting competitive binding of BPA to SBPs in vivo. SBPs carry steroid hormones and endocrine disruptors to steroid-sensitive target tissues and aid in eliciting signal transduction. Furthermore, BPA has positively correlated with increased

128 circulating steroid hormone binding globulin (SHBG) in men (364). As the brain and the hypothalamus are major targets of estrogen, it is likely that SBPs and SHBGs may carry concentrated amounts of endocrine disruptors, including BPA, to the hypothalamus, leading to exposure levels higher than reported urine concentrations. These steroid binding proteins are absent in the in vitro environment, which makes direct comparisons between serum/urine levels and cell culture concentrations impossible. Regardless, the treatment concentration of 100 µM is higher than the amounts reported in vivo and since extracellular actions of these higher concentrations of BPA cannot be completely ruled out, the high concentrations used represent a primary limitation of this study. siRNA knockdown or knockout models were used to confirm the transcription factors involved in BPA-mediated effects on Npy and Agrp; however, inhibitors and receptor antagonists were used for upstream signaling pathways. Inhibitors and antagonists can have non-specific actions against multiple signaling proteins and/or receptors and represents a limitation. While positive controls were used to ensure the chemical had effects in the cell lines (Appendix B), direct inhibition of phosphorylation of signaling proteins or receptor activity was not measured. Importantly, the JNK inhibitor SP600125 was shown to block BPA-mediated changes in Npy expression in the mHypoA-59 and mHypoE-46 cells, but not mHypoE-41 cells (Appendix D). While it is likely that it is JNK that is involved in the regulation of Npy, SP600125 can also reduce oxidative stress (365) and inhibit CAMKKß and AMPK (366). An siRNA targeting JNK1/2 was used, but adequate knockdown of the protein was not achieved as assessed by Western blotting. Compound C was used to inhibit AMPK; however, it is also known to inhibit bone morphogenic protein (367) and alter circadian gene Bmal1 independently of AMPK in mHypoE-37 cells (317). As BMAL1 is known to regulate Npy, this may introduce a confounding variable in the use of compound C. While the concentration of compound C used in our study was previously shown to decrease AMPK phosphorylation in a hypothalamic cell line (317), it was not validated in the specific cell lines used in this study. Thus, using an siRNA against AMPK and JNK if possible or measuring the phosphorylation levels of the proteins (as well as the other proteins targeted by the inhibitors) in the presence of inhibitors can confirm the involvement of these proteins.

Lastly, in chapter 4, the two cell lines displaying a downregulation in Npy were derived from a male embryonic mouse. The upregulation in Npy occurs in cell lines derived from male or female

129 and embryonic or adult mice, illustrating that the upregulation in Npy is not a sex-specific effect. The fact that no female cell lines showed a decrease in Npy expression may be a result of a sex difference in this response or the fact that fewer female-derived cell lines were available to us (only 2 females out of 6 clonal lines tested). This potential sex-specific effect remains to be studied.

6.4 Future Directions

There are several future experiments that can be undertaken to further study the regulation of Npy and Agrp by BPA, but also to determine how the effects of BPA are altered by a combination of other hormonal and environmental factors.

6.4.1 What transcription factors interact with the estrogen receptors to mediate downregulation of Npy in the mHypoE-46 cells?

In Chapter 3 and Chapter 5, ATF3 and BMAL1 were determined as required transcription factors for the upregulation of Agrp and Npy, respectively. ATF3 was also required for the upregulation of Npy in the mHypoE-41 cells. In the mHypoE-46 cells, the reversal of the downregulation in Npy to an upregulation in the presence of estrogen receptor antagonists against ERß and GPER implicates these receptors in BPA-mediated Npy downregulation. However, it remains to be determined which transcription factors mediate the ability of BPA to act via ERß and GPER in the mHypoE-46 cells. As mentioned in chapter 4, one plausible candidate is HSPB1, which can act as a transcriptional cofactor with ERß (296). The lower expression of HSPB1 in the mHypoE-46 cells, compared to the mHypoA-59 and mHypoE-41 cells, leads to the hypothesis that HSPB1 may prevent any ERß-mediated actions in the mHypoA-59 cells. To test this hypothesis, siRNA-mediated knockdown of HSPB1 in the mHypoA-59 cells can be performed followed by a BPA-treatment to determine if the direction of Npy regulation is altered. If so, an ERß antagonist or an siRNA against ERß can be used to determine whether the change in direction can then be reversed by blocking estrogen signaling. In the mHypoE-46 cells, HSPB1 can be overexpressed prior to BPA treatment to determine whether HSPB1 alone can prevent BPA-mediated downregulation of Npy. Finally, chromatin immunoprecipitation can be performed to determine whether ERß binding to the Npy promoter is altered by BPA treatment and whether this effect is specific to the mHypoE-46 cells.

130

6.4.2 Do microRNAs play a role in BPA-mediated Npy regulation?

In Chapter 3, inhibition of transcription with actinomycin D prevented BPA-mediated Agrp and Atf3 upregulation. Furthermore, pre-Agrp mRNA levels were upregulated by BPA, suggesting de novo transcription of Agrp. Finally, translational inhibition using cycloheximide prevented Agrp induction in mHypoA-59 cells (Appendix G), suggesting transcription and translation of a factor (i.e. ATF3) is necessary for the increase in Agrp mRNA. Interestingly, while actinomycin D blocked BPA-mediated increases in Npy in both mHypoA-59 and mHypoE-41 cells, cycloheximide pre-treatment did not completely block the Npy upregulation (Appendix F and G). This raises the question of why transcription of a factor may be required, but not translation. The results from Chapter 5, demonstrating BMAL1 binding to the regulatory region of Npy, may imply that only increased binding and not increased translation of BMAL1 is required to upregulate Npy transcription.

An alternate mechanism whereby transcribed material may influence gene expression without de novo translation of a factor involves regulation by small RNA molecules. microRNAs are small non-coding RNA molecules approximately 22 nt in length that bind to specific mRNA molecules preventing translation of the mRNA into protein or targeting the mRNA for degradation (368). These and other small RNAs can be transcribed by the cell, and subsequently affect the stability of the Npy mRNA transcript without the need for translation of a protein. Indeed, microRNAs have been associated with obesity, insulin resistance and diabetes (369) and BPA has been shown to alter microRNAs in breast carcinoma cells (370), placental cells (371) and in the developmentally-exposed hypothalamus (372). As such, the effect of BPA on Npy expression may include this additional layer of post-transcriptional regulation that can be investigated using microRNA mimics and antagomirs to modulate their levels after BPA treatment.

6.4.3 Does BPA alter AgRP and NPY protein levels?

The experiments in this thesis focussed on changes in Agrp and Npy mRNA expression. Changes in AgRP or NPY protein expression or secretion in response to compounds, such as glucose (139), leptin (123, 373, 374) and estrogen (115, 117), have previously shown correlations to changes in Agrp or Npy mRNA levels. Furthermore, BPA increased the expression of prohormone convertase 2 in hypothalamic neurons (203), suggesting that processing of ProNPY to NPY (375) or post-translational cleavage of AgRP (376) may be

131 increased with BPA treatment. As the peptides are ultimately the functional molecules, analyzing changes in the secretion and/or protein content of AgRP and NPY in response to BPA with an ELISA or immunocytochemistry is an important future direction.

6.4.4 Does BPA lead to hypothalamic resistance to hormones?

The hypothalamus plays an essential role in responding to peripheral signals that are indicative of whole-body energy homeostasis. As mentioned, insulin, leptin and E2 are anorexigenic signals that downregulate Npy and Agrp expression and/or secretion (80, 117). Considering BPA led to a decrease in the expression of Esr1 and Esr2 in this study and has been shown to lead to insulin resistance in peripheral cells (63, 66), whether BPA can induce hypothalamic resistance to these anorexigenic signals merits study. Cells can be pre-treated with varying concentrations of BPA, followed by treatment with insulin, leptin and estrogen. Western blot and secretion assays can be used to determine whether common signaling proteins downstream of these hormones are still activated (pAKT, pSTAT3, pCREB) and whether the downregulation of NPY and/or AgRP secretion persists in the presence of BPA. An inability of BPA-exposed cells to respond to these peripheral cues may exacerbate BPA-induced hypothalamic dysregulation in vivo.

6.4.5 Combination treatments with BPA alternatives and fatty acids

In the current environment, humans and wildlife are not only exposed to BPA. Several other endocrine disrupting chemicals, including, but not limited to, structurally similar BPA alternatives, such as bisphenol S (BPS) and bisphenol F, are widely used in industry as they are thought to be safer (188). However, BPS is emerging as an obesogen and has been shown to upregulate adipogenic gene expression and adipogenesis in both murine and human pre- adipocytes (43, 377). In Swiss albino mice, BPS was recently shown to increase food intake and weight gain with a corresponding increase in Agrp expression (378). Preliminary results from our laboratory show that BPS increases Agrp expression in mHypoE-41, mHypoA-59 and mHypoE- 46 cells. The mechanistic basis of these effects warrants future investigation using similar methods that were used in this thesis.

Besides the bisphenols, several other environmental chemicals, including phthalates, parabens, pesticides and UV-filters are all endocrine disruptors. Many of these chemicals act in

132 combination and mixtures of these chemicals are now being tested. While individually, they may not produce effects at low doses, the mixtures lead to developmental and reproductive abnormalities in rats, such as increased nipple retention and reduced prostate weight (379), as well as advanced neurogenesis in the neuroendocrine hypothalamus of offspring (380). As such, the effect of these mixtures on the hypothalamic regulation of food intake and the Npy/Agrp system is an important area of future investigation.

Finally, the effect of HFDs on the hypothalamic Npy/Agrp system has been investigated in our lab. Recent reports suggest that the dietary saturated fat, palmitate, exacerbates the EndR stress- inducing effects of BPA in mouse embryonic fibroblasts (208). Whether BPA and palmitate have additive effects in hypothalamic neurons can be studied by pre-treating or co-treating with palmitate alongside BPA treatment. The relevance of these studies is undeniable as the majority of western society is concurrently exposed to both of these obesogens.

6.4.6 Can antioxidant supplements protect from BPA-induced weight gain?

One of the most intriguing results from these studies is the ability of NAC and vitamin B6 to protect against BPA-mediated Npy upregulation in the mHypoA-59 cells. Vitamin B6 has been previously shown to protect against BPA-induced glucose intolerance in mice (288). C57BL6 mice have increased weight gain within 2 weeks of BPA exposure compared to vehicle (240). Although vitamin B6 did not protect against Agrp dysregulation (Appendix C), vitamin B6 or NAC supplementation can be given to mice orally to determine whether it is sufficient to protect against BPA-induced weight gain.

6.5 Conclusions

Overall, these studies demonstrate that hypothalamic Npy and Agrp are susceptible to BPA- induced dysregulation. This suggests that BPA-induced reproductive dysfunction and obesity likely involves alterations in hypothalamic physiology, ultimately leading to changes in energy homeostasis. While AMPK activation is a likely upstream mediator of the effects on Agrp and Npy upregulation and the transcription factor ATF3 is essential for Agrp upregulation, the downstream mediators of Npy regulation are population specific and involve oxidative stress and BMAL1, ATF3, or the estrogen receptors, GPER and ERß. The distinct pathways activated by BPA to regulate Agrp and Npy in the different cell subpopulations speaks to the promiscuity of

133

BPA and other industrial endocrine disruptors. Unlike pharmaceuticals that are designed to target one or two pathways, these have the ability to concurrently disrupt a wide array of systems. Thus, multiple pathways need to be therapeutically targeted in order to prevent or reverse these effects, shedding light on the importance of thorough risk assessment and government policy (24) to limit the use of these chemicals and their replacements.

Appendices Appendix A: Steroid receptor antagonists do not block the BPA mediated increase in Agrp expression At least one cell line out of the mHypoA-59 and mHypoE-41 cells were pre-treated with antagonists or inverse agonists of ERß, GPER, ERR훾, AhR, GR or PPAR, or with an Esr1 siRNA prior to treatment with 100 µM BPA for 8 hours. BPA-induced changes in Agrp expression persisted in the presence of these antagonists. BPA-induced changes in Npy expression in mHypoE-41 cells also persisted in the presence of the GPER antagonist, the ERR훾 inverse agonist and the AhR antagonist.

A mHypoA-59: Agrp 3

+ ** A

N R 2 *** ** m *** ** * * 7 ** * l p=0.0573 * p *

R /

p 1

r g

A

0

1 7 6 6

P

O O O O O O

5 5

M

M M

9 0 8 8

% % % % % %

M

P

M

M

S S S S S S

1 1

µ

µ µ 1 9

4 4

1 1 1 1 1 1

M

µ

T

µ

n

......

3 0

M M M M M M

G G

0

µ U U

1 5

H

0 0 0 0 0 0 1

5

2

7

0

1

D D D D D D

R R

0

P

2

0

0

1

0

1

H

T C GPER ERβ AhR PPARγ GR B mHypoA-59: Agrp

2 Legend A **

N * mHypoA-59 R

m Veh

7 l 1 100 µM BPA

p mHypoE-41

R

/ p

r Veh

g 100 µM BPA A

0 NC Esr1 siRNA

134 135

C mHypoE-41: Agrp D mHypoE-41: Npy ++++

3 15 ####

**** **** A

*** A N

**** N R 2 ** * 10R

m **

*** m

7

7

l

l

p

p

R R

/ **

/ *** p

1 y 5 **

r p

g * *

N

A

0 0

2 2

1 2 2 1

O O O O O O

5

5

M

M

8 8 9 8 8

9

% % %

% % %

S S S

S S S

1 1

M M

µ 1 1 1 1

1 µ 1

1 1 1 1 1 1

M

M

M

M

. . .

. . .

µ µ

5 5 5 5

3 3

M M M M M M

G

G

µ µ

µ

µ

1

1

0 0 0

0 0 0

2

2

K K

K K

0 0

D D D

D D D

0 0

1

1

2 2

1

S S S 1 S

1

1

H H

G G

G G

C C GPER ERRγ AhR GPER ERRγ AhR

Figure A-A: Steroid receptor antagonists do not block the BPA-mediated increase in Agrp expression (A) mHypoA-59 cells or (C, D) mHypoE-41 cells pre-treated with 0.1% DMSO or indicated concentration of receptor antagonist for 1 hour prior to treatment with 100 µM BPA or vehicle (0.05% EtOH) for 8 hours (n=3-5). (B) mHypoA-59 cells transfected with an Esr1 siRNA for 24 hours, followed by treatment with 100 µM BPA or vehicle for 8 hours (n=4). Gene expression was analyzed using qRT-PCR. Data are expressed as mean +/- SEM, and statistical significance was determined using a Two-way ANOVA followed by the Tukey multiple comparison test; Veh vs. BPA: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; DMSO vs. antagonist: ####P<0.0001; interaction: +P<0.05, ++++P<0.0001.

136

Appendix B: Steroid receptor antagonist and PBA positive controls

The estrogen receptor antagonists and the glucocorticoid receptor antagonist have previously been used in hypothalamic neuronal cell lines in our lab to block the effects of estradiol on Kiss1 expression (221) and dexamethasone on RF-amide-related peptide-3 expression (381), respectively. These demonstrate that these receptor antagonists block agonist-mediated actions in hypothalamic cell lines. These experiments do not establish whether these antagonists are capable of blocking BPA-induced effects. Furthermore, AhR and ERR do not have endogenous ligands, and the effect of the activation of these receptors on Npy or Agrp expression is not known. As such, to determine whether the receptors under question were being targeted in the experiment, we measured the BPA-induced responses of genes known to be targeted by the receptor in question in the presence of the specific antagonist. The BPA-mediated decrease in GPER-target Esrr (382) and ER-target Igf1 (222) were mitigated with G15 and PHTPP, respectively. The BPA-mediated decrease in Ahr and ERR-target Pdk4 (223) were mitigated by CH223191 and GSK5182, respectively. BPA-mediated decrease in GR-responsive gene Egr1 (224) was reversed by RU486. Finally, PPAR-antagonist T0090709 has been previously shown to block BPA-mediated increase in Pomc expression in a hypothalamic cell line at the same concentrations of BPA and T0070907 used in this study (203). These positive controls verified that the antagonists do have the potential to block BPA-mediated effects at the concentrations used in this study. PBA has previously been used to block EndR stress in hypothalamic neurons (237). The BPA-mediated increase in Grp78 mRNA expression was abolished in the presence of PBA in both the mHypoA-59 and mHypoE-41 cells, indicating that PBA was able to block EndR stress.

137

A Steroid receptor antagonists i. GPR30 ii. ERβ iii. ERRγ

2.0 2.0 2.0 ++++

A Steroid receptor antagonists

A A

A ns

N N

1.5 N 1.5 1.5

R

R

R m

ns m

m

7

7

l

l

7

l p

1.0 1.0 p 1.0 ##

p

R

R /

/ R

** /

α

4

r 1

* k ****

f r

0.5 0.5 d 0.5 s

g

I P E 0.0 0.0 0.0 DMSO 5 µM DMSO 5 µM DMSO 10 µM G15 PHTPP GSK5182 iv. AhR v. GR +++ 1.5 2.0 + 2.5

p=0.06 ##

A A A **

N 2.0 N

N 1.5

R R 1.0 R

* m

m

m

1.5

7

7

7

l l

l 1.0

p p

** p

R R R

/ 1.0

/ /

0.5 1

R

R

r

h h

0.5 g

A A

E 0.5

0.0 0.0 0.0 DMSO 10 µM DMSO 10 µM DMSO 10 µM CH223191 CH223191 RU486 B PBA B PBA mHypoA-59 mHypoE-41 mHypoA-59 mHypoE-41 ++ ++ Legend ++ ++++ 3 3 ++ mHypoA-59

++++ A A *** Veh N ***

N 100 µM BPA

R R

m 2

m 2 **** mHypoE-41

****

7

l

7 l p ## ## Veh p ### ## R ## ### / ### 100 µM BPA

R ### 8 / 1

7 8

1 p

7

r p

G r

G 0 0 H2O 5 mM 10 mM H2O 5 mM 10 mM PBA PBA PBA PBA H2O 5 mM 10 mM H2O 5 mM 10 mM PBA PBA PBA PBA

Figure A-B: Steroid receptor antagonist and PBA positive controls mHypoA-59 cells (blue) or mHypoE-41 cells (pink) were pre-treated with DMSO or the indicated concentrations of the receptor antagonists (A) or H2O or PBA (B) for 1 hour, followed by treatment with 0.05% EtOH (Veh) or 100 M BPA for 8 hours (n=3-4). Gene expression was analyzed using qRT-PCR. Data are expressed as mean +/- SEM, and statistical significance was determined using a Two-way ANOVA followed by the Tukey multiple comparison test; Veh vs. BPA: *P<0.05, **P<0.01, ****P<0.0001; DMSO vs. antagonist: ##P<0.01; interaction: +P<0.0.05, ++++P<0.0001

138

Appendix C: The BPA-induced increase in Agrp expression is not mediated by oxidative stress.

The protective effects of NBA media against BPA-induced changes in Npy expression in the mHypoA-59 and mHypoE-41 cells were remarkable. Interestingly, this effect was specific to Npy, and the upregulation in Agrp in the mHypoA-59 cells was unchanged in the presence of NBA media. Similarly, none of TUDCA, JNK inhibition, NDGA nor vitamin B6 prevented the changes in Agrp expression in the mHypoA-59 cells. Furthermore, although NAC may prevent the upregulation of Agrp in the mHypoA-59 cells (only n=2), it did not prevent the upregulation in the mHypoE-41 cells, illustrating that BPA-induced changes in Agrp expression is generally independent of the induction of oxidative stress.

A mHypoA-59: Agrp 3

A * *** N ***

R 2 ** **

m **

7 l ** * p=0.053

p *

R

/ p

r 1

g

A

0

5

A

A

A

O O

M O O O M

M

4

M

M

M M

% %

2 2 2

B

6

µ

S S C

G

E

C

6

1

µ

µ µ

1 2

m

m

B

N H H H

. .

A

1

D 0

D

B

M M

M

0

4

0

5 0 0

0 2

S

N

. N

U

D D D 4

2

1

2 P T

B mHypoE-41: Agrp Legend mHypoA-59 Veh ++ 100 µM BPA 3 # mHypoE-41 ## ** *** Veh ** 100 µM BPA

A *** N

R 2 ** ** m **** ***

7 **

l

p

R

/ p

r 1

g

A

0

A A

A

M M O O O M

M

M

M

2 2 2

B B

6

µ

C

E E

C

6

µ

m

m

B

N N H H H

A

0

D

B

M M

0

5

0 2

N

.

U

D D 4

1

2 T 8 hours 16 hours

139

Figure A-C: BPA-mediated change in Agrp expression is not mediated by oxidative stress (A) mHypoA-59 or (B) mHypoE-41 cells pre-treated with TUDCA, NDGA, PS1145 or NAC for 1 hour or 16 hours (NAC) or co-treated in NBA or with vitamin B6, alongside an 8 or 16 hour (N41 as indicated) treatment with 100 M BPA or vehicle (0.05% EtOH). Agrp expression was analyzed using qRT-PCR (n=3-5, except A59 NAC n=2). Data are expressed as mean +/- SEM, and statistical significance was determined using a Two-way ANOVA followed by the Tukey multiple comparison test; Veh vs. BPA: *P<0.05, **P<0.01, ***P<0.001; Veh vs. pre-/co- treatment: #P<0.05, ##P<0.01; interaction: ++P<0.01.

140

Appendix D: The effect of BPA on Npy expression in the mHypoE- 41 cells is dependent on ATF3 rather than oxidative stress

Unlike the mHypoA-59, the mechanism underlying Npy regulation by BPA in the mHypoE-41 cells was more complex. Interestingly, the BPA-mediated upregulation in Npy in the mHypoE-41 cells was blocked in the presence of an ATF3 siRNA. This was in contrast to the mHypoA-59 cells, where the ATF3 siRNA downregulated basal Npy levels, but did not affect the BPA- mediated increase in Npy (Fig 3-6). Thus, it is clear that the mechanisms are not identical in A mHypoE-41: Npy these cell lines in terms of Npy transcriptional upregulation. In support of this phenomena, not all 15 ++

compounds that blocked the increase in Npy in the mHypoA-59 cells were able to prevent the +++ A

N ****

R ####

upregulation in the mHypoE-41 cells. NBA media and TUDCA did not block the 10increase in Npy ## m

****

7 **** ++ l

by BPA in mHypoE-41 cells. Furthermore, NAC and vitamin B6 had only a partialp effect to**** # R *** **** / *** **** y 5 ** *** **** p ****

diminish the magnitude of Npy upregulation in the mHypoE-41 cells. However, preN -treatment **

with CC completely blocked the BPA-mediated increase in Npy in the mHypoE-410 cells. As

A A

A

O

M M O M O O M

M

M

M

%

2 2 2

B B

6

P

µ µ

S C

E E

C

6

µ

1

m

m

B

N N H S H H

.

A

0 0

D

B

M

M M

0

0

5

5 0 2

N

.

U

D D 4

such, in the mHypoE-41 cells, oxidative stress does not entirely mediate BPA-induced NpyD

1

2 T regulation, and requires ATF3, likely activated by AMPK, as well. 8 hours 16 hours A mHypoE-41: Npy B mHypoE-41: Npy Veh 15 100 µM BPA ++ ++ 3 +++

A *** A

N **** N

R ####

10 ## R 2

m m

****

7 **** ++

l

7 l

p **** # p

R ***

**** R / *** **** / ### y 5 ***

** **** y 1 p **** p

N ** N

0

0

A

A

A

O

M

M

O

M

O

O M

M 0.1% 6 µM

M

M

%

2

2

2

B

B

6

P

µ

µ

S

C

E

E

C

6

µ

1

m

m

B

N

N

H

S

H

H

.

A

0

0

D

B

M M

M DMSO CC

0

0

5

5

0

2

N

.

U

D

D

D

4

1

2 T 8 hours 16 hours

B mHFigureypoE-4 1A: Np-D:y Antioxidants do not prevent BPA-induced Npy expression in the mHypoE-41 cells. (A) mHypoE++ -41 cells pre-treated with TUDCA, SP600125 (SP) or NAC for 1 or 16 hours (NAC) 3 or co-treated in NBA or with vitamin B6, alongside an 8 or 16 hour (for NBA as indicated) ***

A treatment with 100 M BPA or vehicle (0.05% EtOH). (B) mHypoE-41 cells pre-treated with CC N

R 2 for 1 hour, followed by treatment with 100 M BPA or vehicle (0.05% EtOH) for 8 hours. Npy

m

7 expression was analyzed using qRT-PCR. Data are expressed as mean +/- SEM, and statistical

l p

R significance was determined using a Two-way ANOVA followed by the Tukey multiple

/ ###

y 1

p comparison test; Veh vs. BPA: **P<0.01, ***P<0.001, ****P<0.0001; Veh vs. pre-/co-treatment: N #P<0.05, ##P<0.01, ###P<0.001, ####P<0.0001; interaction: ++P<0.01, +++P<0.001. 0 0.1% 6 µM DMSO CC

141

Appendix E: The effect of BPA on Npy in primary culture

In CD1 male and female derived hypothalamic primary culture, treatment with 100 µM BPA did not alter Npy expression at 8 hours.

A CD1 Male Primary Culture: Npy 1.5 Veh

A 100 µM BPA

N R

m 1.0

7

l

p

R /

0.5

y

p N 0.0 NBA DMEM

Figure A-E: BPA does not increase Npy mRNA expression in hypothalamic primary culture at 8 hours. (A) Primary culture derived from the hypothalamus of male CD-1 mice treated with 100 µM BPA or vehicle for 8 hours (n=4) in NBA or DMEM media. Npy expression was analyzed using qRT-PCR and data are represented as mean +/- SEM.

142

Appendix F: Transcriptional inhibition affects the BPA-mediated changes in Npy

In both mHypoA-59 and mHypoE-41 cells, inhibition of transcription using 10 µg/ml ActD prevented the BPA-mediated increase in Npy expression. While this does not imply direct transcriptional upregulation of Npy as ActD inhibits global transcription, it suggests that either de novo Npy transcription occurs and/or that transcription of some factor is necessary for the BPA- mediated Npy increase.

A Transcriptional Inhibition: Npy ++++

6 6

A **** A

+

N

N R R Legend

4 4 m

m mHypoA-59

7

7 l

l * Veh p

p 100 µM BPA

R

R /

2 / 2 mHypoE-41 y

y Veh p p ##

#### 100 µM BPA

N N 0 0 DMSO ActD DMSO ActD Pretreatment Pretreatment

Figure A-F: Transcriptional inhibition blocks the BPA-mediated increase in Npy in mHypoA-59 and mHypoE-41 cells. (A) mHypoA-59 and mHypoE-41 cells pre-treated with 0.1% DMSO or 10 µg/ml ActD for 1 hour, followed by treatment with 100 µM BPA or vehicle (0.05% EtOH) for 8 hours. Npy expression was analyzed using qRT-PCR. Data are expressed as mean +/- SEM, and statistical significance was determined using a Two-way ANOVA followed by the Tukey multiple comparison test; Veh vs. BPA: *P<0.05,****P<0.0001; DMSO vs. ActD: ##P<0.01, ####P<0.0001; interaction: +P<0.05, ++++P<0.0001.

143

Appendix G: Translational inhibition affects the BPA-mediated induction of BPA of Npy and Agrp

In the mHypoA-59 cells, translational inhibition with 5 µM cycloheximide (Cyc) increased basal levels of Npy, but did not prevent the BPA-mediated increase in Npy expression. Regardless, the magnitude of Npy induction by BPA showed a decreased trend in the Cyc pre-treated group (10- fold induction in DMSO-treated group versus ~2-fold induction in 5 µM cycloheximide-treated group; similar pattern with 35 µM Cyc). The increase in Agrp expression was diminished with 5 µM Cyc and significantly inhibited with 35 µM Cyc pre-treatment. These results suggest that translation of a factor is likely involved in the BPA-mediated upregulation of Agrp and partially involved in the upregulation of Npy.

A Translational Inhibition: Npy B Translational Inhibition: Agrp

p=0.0519

#### +

40 **** 2.0 A

A * N

N 30 1.5

R

R

m

m

####

7

l 7 l 1.0

20 p

p

R

R

/ /

** p

y

r p

10 g 0.5

N A

0 0.0 0.1% 5 µM 0.1% 5 µM 35 µM DMSO Cyc DMSO Cyc Cyc

Veh 100 µM BPA

Figure A-G: Translational inhibition blocks the BPA-mediated increase in Agrp, but not Npy. (A, B) mHypoA-59 cells pre-treated with 0.1% DMSO, 5 µM or 35 µM cycloheximide (Cyc) for 1 hour, followed by treatment with 100 µM BPA or vehicle (0.05% EtOH) for 8 hours. Gene expression was analyzed using qRT-PCR. Data are expressed as mean +/- SEM, and statistical significance was determined using a Two-way ANOVA followed by the Tukey multiple comparison test; Veh vs. BPA: P<0.05,*P<0.01, ****P<0.0001; DMSO vs. Cyc: ####P<0.0001; interaction: +P<0.05. 35 µM Cyc was used to analyze translational inhibition for Npy expression as well, and BPA-mediated Npy induction was still observed.

144

Appendix H: Esr2 expression is decreased in mHypoA-BMAL1- KO cells

The mRNA expression of Esr2 in mHypoA-BMAL1-KO/F and /M cells is approximately 1/3 of the levels present in mHypoA-BMAL1-WT/F cells, suggesting lower levels of ERß in the absence of BMAL1.

2.0

p=0.07 A

N 1.5

R ***

m

7

l 1.0

p

R

/

2 r

s 0.5 E

0.0 WT KO/F WT KO/M mHypoA-Bmal1-WT/F-Vehicle mHypoA-Bmal1-KO/F -Vehicle mHypo-Bmal1-KO/M-Vehicle Figure A-H Esr2 mRNA levels are lower in mHypoA-Bmal1-KO cells compared with mHypoA-Bmal1-WT/F cells. mHypoA-Bmal1-KO/F or /M cells treated with 0.05% EtOH for 8 hours alongside mHypoA- Bmal1-WT/F cells. Esr2 mRNA levels were analyzed using qRT-PCR (n=3-5). Data are presented as mean +/- SEM and statistical significance was determined using a student T-test, ***P<0.001.

References

1. World Health Organization. Obesity and Overweight. 2020; https://www.who.int/news- room/fact-sheets/detail/obesity-and-overweight. Accessed September 1, 2020.

2. Mitchell NS, Catenacci VA, Wyatt HR, Hill JO. Obesity: overview of an epidemic. Psychiatr Clin North Am 2011; 34:717-732

3. World Health Organization. Global Health Observatory (GHO) data: Mean Body Mass Index (BMI). 2008; https://www.who.int/gho/ncd/risk_factors/bmi_text/en/. Accessed September 1, 2020.

4. World Health Organization. Sexual and reproductive health:Infertility is a global public health issue. 2020; https://www.who.int/reproductivehealth/topics/infertility/perspective/en/. Accessed September 20, 2020.

5. Health Canada. Fertility. 2019; https://www.canada.ca/en/public- health/services/fertility/fertility.html. Accessed September 1, 2020.

6. World Health Organization. Sexual and reproductive health: Global prevalence of infertility, infecundity and childlessness. 2004; https://www.who.int/reproductivehealth/topics/infertility/burden/en/. Accessed September 20, 2020.

7. Deshpande PS, Gupta AS. Causes and Prevalence of Factors Causing Infertility in a Public Health Facility. J Hum Reprod Sci 2019; 12:287-293

8. Winsor SP, Ala-Leppilampi K, Spitzer K, Edney DR, Petropanagos A, Cadesky KI, Casper R, Laskin C. The Affordability and Accessibility of Ontario's Publicly Funded IVF Program: A Survey of Patients. J Obstet Gynaecol Can 2020; 42:568-575

9. Silvestris E, de Pergola G, Rosania R, Loverro G. Obesity as disruptor of the female fertility. Reprod Biol Endocrinol 2018; 16:22

10. Sallmen M, Sandler DP, Hoppin JA, Blair A, Baird DD. Reduced fertility among overweight and obese men. Epidemiology 2006; 17:520-523

11. Katib A. Mechanisms linking obesity to male infertility. Cent European J Urol 2015; 68:79-85

12. Baillie-Hamilton PF. Chemical toxins: a hypothesis to explain the global obesity epidemic. J Altern Complement Med 2002; 8:185-192

13. Zoeller RT, Brown TR, Doan LL, Gore AC, Skakkebaek NE, Soto AM, Woodruff TJ, Vom Saal FS. Endocrine-disrupting chemicals and public health protection: a statement of principles from The Endocrine Society. Endocrinology 2012; 153:4097-4110

145 146

14. Gore AC, Chappell VA, Fenton SE, Flaws JA, Nadal A, Prins GS, Toppari J, Zoeller RT. EDC-2: The Endocrine Society's Second Scientific Statement on Endocrine-Disrupting Chemicals. Endocrine Reviews 2015; 36:E1-E150

15. Diamanti-Kandarakis E, Bourguignon JP, Giudice LC, Hauser R, Prins GS, Soto AM, Zoeller RT, Gore AC. Endocrine-disrupting chemicals: an Endocrine Society scientific statement. Endocr Rev 2009; 30:293-342

16. Arora S, Anubhuti. Role of neuropeptides in appetite regulation and obesity--a review. Neuropeptides 2006; 40:375-401

17. Acosta-Martinez M, Horton T, Levine JE. Estrogen receptors in neuropeptide Y neurons: at the crossroads of feeding and reproduction. Trends Endocrinol Metab 2007; 18:48-50

18. Tollefsen KE. Interaction of estrogen mimics, singly and in combination, with plasma sex steroid-binding proteins in rainbow trout (Oncorhynchus mykiss). Aquat Toxicol 2002; 56:215-225

19. Depypere HT, Bolca S, Bracke M, Delanghe J, Comhaire F, Blondeel P. The serum estradiol concentration is the main determinant of the estradiol concentration in normal breast tissue. Maturitas 2015; 81:42-45

20. La Merrill MA, Vandenberg LN, Smith MT, Goodson W, Browne P, Patisaul HB, Guyton KZ, Kortenkamp A, Cogliano VJ, Woodruff TJ, Rieswijk L, Sone H, Korach KS, Gore AC, Zeise L, Zoeller RT. Consensus on the key characteristics of endocrine- disrupting chemicals as a basis for hazard identification. Nat Rev Endocrinol 2020; 16:45-57

21. Welshons WV, Nagel SC, vom Saal FS. Large effects from small exposures. III. Endocrine mechanisms mediating effects of bisphenol A at levels of human exposure. Endocrinology 2006; 147:S56-69

22. Ashby J, Odum J, Paton D, Lefevre PA, Beresford N, Sumpter JP. Re-evaluation of the first synthetic estrogen, 1-keto-1,2,3, 4-tetrahydrophenanthrene, and bisphenol A, using both the ovariectomised rat model used in 1933 and additional assays. Toxicol Lett 2000; 115:231-238

23. Vogel SA. The politics of plastics: the making and unmaking of bisphenol a "safety". Am J Public Health 2009; 99 Suppl 3:S559-566

24. Kassotis CD, Vandenberg LN, Demeneix BA, Porta M, Slama R, Trasande L. Endocrine- disrupting chemicals: economic, regulatory, and policy implications. Lancet Diabetes Endocrinol 2020; 8:719-730

25. Vandenberg LN, Hauser R, Marcus M, Olea N, Welshons WV. Human exposure to bisphenol A (BPA). Reprod Toxicol 2007; 24:139-177

26. Yamamoto T, Yasuhara A, Shiraishi H, Nakasugi O. Bisphenol A in hazardous waste landfill leachates. Chemosphere 2001; 42:415-418

147

27. Gewurtz SB, Tardif G, Power M, Backus S, Dove A, Dube-Roberge K, Garron C, King M, Lalonde B, Letcher RJ, Martin PA, McDaniel TV, McGoldrick DJ, Pelletier M, Small J, Smyth SA, Teslic S, Tessier J. Bisphenol A in the Canadian Environment: A Multimedia Analysis. Science of The Total Environment: Elsevier; 2020.

28. Mohapatra DP, Brar SK, Tyagi RD, Surampalli RY. Occurrence of bisphenol A in wastewater and wastewater sludge of CUQ treatment plant. Vol 12011:3-3.

29. Wang H, Liu ZH, Zhang J, Huang RP, Yin H, Dang Z. Human exposure of bisphenol A and its analogues: understandings from human urinary excretion data and wastewater- based epidemiology. Environ Sci Pollut Res Int 2020; 27:3247-3256

30. Cao XL, Corriveau J, Popovic S. Bisphenol a in canned food products from canadian markets. J Food Prot 2010; 73:1085-1089

31. Kwon BG, Chung SY, Saido K. Sandy beaches as hotspots of bisphenol A. Environ Res 2020:110175

32. Mielke H, Partosch F, Gundert-Remy U. The contribution of dermal exposure to the internal exposure of bisphenol A in man. Toxicol Lett 2011; 204:190-198

33. Marquet F, Payan JP, Beydon D, Wathier L, Grandclaude MC, Ferrari E. In vivo and ex vivo percutaneous absorption of [14C]-bisphenol A in rats: a possible extrapolation to human absorption? Arch Toxicol 2011; 85:1035-1043

34. Hines CJ, Jackson MV, Christianson AL, Clark JC, Arnold JE, Pretty JR, Deddens JA. Air, Hand Wipe, and Surface Wipe Sampling for Bisphenol A (BPA) among Workers in Industries that Manufacture and Use BPA in the United States. Journal of Occupational and Environmental Hygiene 2017:00-00

35. Bushnik T, Haines D, Levallois P, Levesque J, Van Oostdam J, Viau C. Lead and bisphenol A concentrations in the Canadian population. Health Rep 2010; 21:7-18

36. Gerona R, Vom Saal FS, Hunt PA. BPA: have flawed analytical techniques compromised risk assessments? Lancet Diabetes Endocrinol 2020; 8:11-13

37. Charisiadis P, Andrianou XD, van der Meer TP, den Dunnen WFA, Swaab DF, Wolffenbuttel BHR, Makris KC, van Vliet-Ostaptchouk JV. Possible Obesogenic Effects of Bisphenols Accumulation in the Human Brain. Sci Rep 2018; 8:8186

38. van der Meer TP, Artacho-Cordón F, Swaab DF, Struik D, Makris KC, Wolffenbuttel BHR, Frederiksen H, van Vliet-Ostaptchouk JV. Distribution of Non-Persistent Endocrine Disruptors in Two Different Regions of the Human Brain. Int J Environ Res Public Health 2017; 14

39. Stahlhut RW, Welshons WV, Swan SH. Bisphenol A data in NHANES suggest longer than expected half-life, substantial nonfood exposure, or both. Environ Health Perspect 2009; 117:784-789

148

40. Nishikawa M, Iwano H, Yanagisawa R, Koike N, Inoue H, Yokota H. Placental transfer of conjugated bisphenol A and subsequent reactivation in the rat fetus. Environ Health Perspect 2010; 118:1196-1203

41. Boucher JG, Boudreau A, Ahmed S, Atlas E. In Vitro Effects of Bisphenol A β-D- Glucuronide (BPA-G) on Adipogenesis in Human and Murine Preadipocytes. Environ Health Perspect 2015; 123:1287-1293

42. Nesan D, Sewell LC, Kurrasch DM. Opening the black box of endocrine disruption of brain development: Lessons from the characterization of Bisphenol A. Horm Behav 2018; 101:50-58

43. Ahmed S, Atlas E. Bisphenol S- and bisphenol A-induced adipogenesis of murine preadipocytes occurs through direct peroxisome proliferator-activated receptor gamma activation. International journal of obesity (2005) 2016;

44. Boucher JG, Boudreau A, Atlas E. Bisphenol A induces differentiation of human preadipocytes in the absence of glucocorticoid and is inhibited by an estrogen-receptor antagonist. Nutrition & diabetes 2014; 4:e102-e102

45. Rancière F, Lyons JG, Loh VHY, Botton J, Galloway T, Wang T, Shaw JE, Magliano DJ. Bisphenol A and the risk of cardiometabolic disorders: a systematic review with meta-analysis of the epidemiological evidence. Environmental health : a global access science source 2015; 14:46

46. Do MT, Chang VC, Mendez MA, de Groh M. Urinary bisphenol A and obesity in adults: results from the Canadian Health Measures Survey. Health Promot Chronic Dis Prev Can 2017; 37:403-412

47. Hong S-h, Sung Y-A, Hong YS, Ha E, Jeong K, Chung H, Lee H. Urinary bisphenol A is associated with insulin resistance and obesity in reproductive-aged women. Clinical Endocrinology 2017; 86:506-512

48. Kim KY, Lee E, Kim Y. The Association between Bisphenol A Exposure and Obesity in Children-A Systematic Review with Meta-Analysis. Int J Environ Res Public Health 2019; 16

49. Teppala S, Madhavan S, Shankar A. Bisphenol A and Metabolic Syndrome: Results from NHANES. Int J Endocrinol 2012; 2012:598180

50. Song Y, Hauser R, Hu FB, Franke AA, Liu S, Sun Q. Urinary concentrations of bisphenol A and phthalate metabolites and weight change: a prospective investigation in US women. Int J Obes (Lond) 2014; 38:1532-1537

51. Health Canada. Health Risk Assessment of Bisphenol A from Food Packaging Applications. 2008; https://www.canada.ca/en/health-canada/services/food- nutrition/food-safety/packaging-materials/bisphenol/health-risk-assessment-bisphenol- food-packaging-applications.html. Accessed November 15, 2020.

149

52. Rubin BS, Murray MK, Damassa DA, King JC, Soto AM. Perinatal exposure to low doses of bisphenol A affects body weight, patterns of estrous cyclicity, and plasma LH levels. Environ Health Perspect 2001; 109:675-680

53. Nikaido Y, Yoshizawa K, Danbara N, Tsujita-Kyutoku M, Yuri T, Uehara N, Tsubura A. Effects of maternal xenoestrogen exposure on development of the reproductive tract and mammary gland in female CD-1 mouse offspring. Reprod Toxicol 2004; 18:803-811

54. Angle BM, Do RP, Ponzi D, Stahlhut RW, Drury BE, Nagel SC, Welshons WV, Besch- Williford CL, Palanza P, Parmigiani S, vom Saal FS, Taylor JA. Metabolic disruption in male mice due to fetal exposure to low but not high doses of bisphenol A (BPA): evidence for effects on body weight, food intake, adipocytes, leptin, , insulin and glucose regulation. Reprod Toxicol 2013; 42:256-268

55. Alonso-Magdalena P, Vieira E, Soriano S, Menes L, Burks D, Quesada I, Nadal A. Bisphenol A exposure during pregnancy disrupts glucose homeostasis in mothers and adult male offspring. Environ Health Perspect 2010; 118:1243-1250

56. Boudalia S, Berges R, Chabanet C, Folia M, Decocq L, Pasquis B, Abdennebi-Najar L, Canivenc-Lavier MC. A multi-generational study on low-dose BPA exposure in Wistar rats: effects on maternal behavior, flavor intake and development. Neurotoxicol Teratol 2014; 41:16-26

57. Mackay H, Patterson ZR, Khazall R, Patel S, Tsirlin D, Abizaid A. Organizational effects of perinatal exposure to bisphenol-A and diethylstilbestrol on arcuate nucleus circuitry controlling food intake and energy expenditure in male and female CD-1 mice. Endocrinology 2013; 154:1465-1475

58. Anderson OS, Peterson KE, Sanchez BN, Zhang Z, Mancuso P, Dolinoy DC. Perinatal bisphenol A exposure promotes hyperactivity, lean body composition, and hormonal responses across the murine life course. FASEB J 2013; 27:1784-1792

59. Newbold RR, Jefferson WN, Padilla-Banks E. Long-term adverse effects of neonatal exposure to bisphenol A on the murine female reproductive tract. Reprod Toxicol 2007; 24:253-258

60. Yang M, Chen M, Wang J, Xu M, Sun J, Ding L, Lv X, Ma Q, Bi Y, Liu R, Hong J, Ning G. Bisphenol A Promotes Adiposity and Inflammation in a Nonmonotonic Dose-response Way in 5-week-old Male and Female C57BL/6J Mice Fed a Low-calorie Diet. Endocrinology 2016; 157:2333-2345

61. Rubin BS, Paranjpe M, DaFonte T, Schaeberle C, Soto AM, Obin M, Greenberg AS. Perinatal BPA exposure alters body weight and composition in a dose specific and sex specific manner: The addition of peripubertal exposure exacerbates adverse effects in female mice. Reproductive Toxicology 2017; 68:130-144

62. Moghaddam HS, Samarghandian S, Farkhondeh T. Effect of bisphenol A on blood glucose, lipid profile and oxidative stress indices in adult male mice. Toxicol Mech Methods 2015; 25:507-513

150

63. Batista TM, Alonso-Magdalena P, Vieira E, Amaral MEC, Cederroth CR, Nef S, Quesada I, Carneiro EM, Nadal A. Short-Term Treatment with Bisphenol-A Leads to Metabolic Abnormalities in Adult Male Mice. PLoS ONE 2012; 7:e33814-e33814

64. Sargis RM, Johnson DN, Choudhury RA, Brady MJ. Environmental Endocrine Disruptors Promote Adipogenesis in the 3T3-L1 Cell Line through Glucocorticoid Receptor Activation. Obesity 2009; 18:1283-1288

65. Wang J, Sun B, Hou M, Pan X, Li X. The environmental obesogen bisphenol A promotes adipogenesis by increasing the amount of 11β-hydroxysteroid dehydrogenase type 1 in the adipose tissue of children. International journal of obesity (2005) 2013; 37:999-1005

66. Alonso-Magdalena P, Morimoto S, Ripoll C, Fuentes E, Nadal A. The estrogenic effect of bisphenol A disrupts pancreatic beta-cell function in vivo and induces insulin resistance. Environ Health Perspect 2006; 114:106-112

67. Soriano S, Alonso-Magdalena P, García-Arévalo M, Novials A, Muhammed SJ, Salehi A, Gustafsson J-A, Quesada I, Nadal A. Rapid insulinotropic action of low doses of bisphenol-A on mouse and human islets of Langerhans: role of estrogen receptor β. PloS one 2012; 7:e31109

68. Ziv-Gal A, Flaws JA. Evidence for bisphenol A-induced female infertility: a review (2007-2016). Fertil Steril 2016; 106:827-856

69. Kandaraki E, Chatzigeorgiou A, Livadas S, Palioura E, Economou F, Koutsilieris M, Palimeri S, Panidis D, Diamanti-Kandarakis E. Endocrine disruptors and polycystic ovary syndrome (PCOS): elevated serum levels of bisphenol A in women with PCOS. J Clin Endocrinol Metab 2011; 96:E480-484

70. Lee SG, Kim JY, Chung JY, Kim YJ, Park JE, Oh S, Yoon YD, Yoo KS, Yoo YH, Kim JM. Bisphenol A exposure during adulthood causes augmentation of follicular atresia and luteal regression by decreasing 17β-estradiol synthesis via downregulation of aromatase in rat ovary. Environ Health Perspect 2013; 121:663-669

71. Castellini C, Totaro M, Parisi A, D'Andrea S, Lucente L, Cordeschi G, Francavilla S, Francavilla F, Barbonetti A. Bisphenol A and Male Fertility: Myths and Realities. Front Endocrinol (Lausanne) 2020; 11:353

72. Li DK, Zhou Z, Miao M, He Y, Wang J, Ferber J, Herrinton LJ, Gao E, Yuan W. Urine bisphenol-A (BPA) level in relation to semen quality. Fertil Steril 2011; 95:625-630 e621-624

73. Chouhan S, Yadav SK, Prakash J, Westfall S, Ghosh A, Agarwal NK, Singh SP. Increase in the expression of inducible nitric oxide synthase on exposure to bisphenol A: A possible cause for decline in steroidogenesis in male mice. Environmental Toxicology and Pharmacology 2015; 39:405-416

151

74. Lombó M, Fernández-Díez C, González-Rojo S, Herráez MP. Genetic and epigenetic alterations induced by bisphenol A exposure during different periods of spermatogenesis: from spermatozoa to the progeny. Sci Rep 2019; 9:18029

75. McIlwraith EK, Belsham DD. Hypothalamic reproductive neurons communicate through signal transduction to control reproduction. Mol Cell Endocrinol 2020:110971

76. Arora S, Anubhuti. Role of neuropeptides in appetite regulation and obesity - A review. Neuropeptides 2006;

77. ANAND BK, BROBECK JR. Hypothalamic control of food intake in rats and cats. Yale J Biol Med 1951; 24:123-140

78. Albert DJ, Storlien LH, Albert JG, Mah CJ. Obesity following disturbance of the ventromedial hypothalamus: a comparison of lesions, lateral cuts, and anterior cuts. Physiol Behav 1971; 7:135-141

79. Dalvi A, Tung S, Belsham DDPS, Nazarians A. Immortalized neurons for the study of hypothalamic function. Am J Physiol Regul Integr Comp Physiol 2011; 300:23-23

80. Morton GJ, Schwartz MW. The NPY/AgRP neuron and energy homeostasis. Int J Obes Relat Metab Disord 2001; 25 Suppl 5:S56-62

81. Zarjevski N, Cusin I, Vettor R, Rohner-Jeanrenaud F, Jeanrenaud B. Chronic intracerebroventricular neuropeptide-Y administration to normal rats mimics hormonal and metabolic changes of obesity. Endocrinology 1993; 133:1753-1758

82. Hulsey MG, Pless CM, White BD, Martin RJ. ICV administration of anti-NPY antisense oligonucleotide: effects on feeding behavior, body weight, peptide content and peptide release. Regulatory peptides 1995; 59:207-214

83. Sahu A, Kalra PS, Kalra SP. Food deprivation and ingestion induce reciprocal changes in neuropeptide Y concentrations in the paraventricular nucleus. Peptides 1988; 9:83-86

84. Hahn TM, Breininger JF, Baskin DG, Schwartz MW. Coexpression of Agrp and NPY in fasting-activated hypothalamic neurons. Nat Neurosci 1998; 1:271-272

85. Sousa-Ferreira L, Garrido M, Nascimento-Ferreira I, Nobrega C, Santos-Carvalho A, Alvaro AR, Rosmaninho-Salgado J, Kaster M, Kugler S, de Almeida LP, Cavadas C. Moderate long-term modulation of neuropeptide Y in hypothalamic arcuate nucleus induces energy balance alterations in adult rats. PLoS One 2011; 6:e22333

86. Hagan MM, Rushing PA, Pritchard LM, Schwartz MW, Strack AM, Van Der Ploeg LH, Woods SC, Seeley RJ. Long-term orexigenic effects of AgRP-(83---132) involve mechanisms other than melanocortin receptor blockade. American journal of physiology Regulatory, integrative and comparative physiology 2000; 279:R47-52

87. Betley JN, Cao ZF, Ritola KD, Sternson SM. Parallel, redundant circuit organization for homeostatic control of feeding behavior. Cell 2013; 155:1337-1350

152

88. Steculorum SM, Ruud J, Karakasilioti I, Backes H, Engström Ruud L, Timper K, Hess ME, Tsaousidou E, Mauer J, Vogt MC, Paeger L, Bremser S, Klein AC, Morgan DA, Frommolt P, Brinkkötter PT, Hammerschmidt P, Benzing T, Rahmouni K, Wunderlich FT, Kloppenburg P, Brüning JC. AgRP Neurons Control Systemic Insulin Sensitivity via Myostatin Expression in Brown Adipose Tissue. Cell 2016; 165:125-138

89. Zhang W, Cline MA, Gilbert ER. Hypothalamus-adipose tissue crosstalk: neuropeptide Y and the regulation of energy metabolism. Nutr Metab (Lond) 2014;

90. Billington CJ, Briggs JE, Grace M, Levine AS. Effects of intracerebroventricular injection of neuropeptide Y on energy metabolism. Am J Physiol 1991; 260:R321-327

91. Stafford JM, Yu F, Printz R, Hasty AH, Swift LL, Niswender KD. Central nervous system neuropeptide Y signaling modulates VLDL triglyceride secretion. Diabetes 2008; 57:1482-1490

92. Rossi M, Kim MS, Morgan DGA, Small CJ, Edwards CMB, Sunter D, Abusnana S, Goldstone AP, Russell SH, Stanley SA, Smith DM, Yagaloff K, Ghatei MA, Bloom SR. A C-Terminal Fragment of Agouti-Related Protein Increases Feeding and Antagonizes the Effect of Alpha-Melanocyte Stimulating Hormone in Vivo. Endocrinology 1998; 139:4428-4431

93. McIlwraith EK, Loganathan N, Belsham DD. Phoenixin Expression Is Regulated by the Fatty Acids Palmitate, Docosahexaenoic Acid and Oleate, and the Endocrine Disrupting Chemical Bisphenol A in Immortalized Hypothalamic Neurons. Front Neurosci 2018; 12:838

94. Wang L, Moenter SM. Differential Roles of Hypothalamic AVPV and Arcuate Kisspeptin Neurons in Estradiol Feedback Regulation of Female Reproduction. Neuroendocrinology 2020; 110:172-184

95. Titolo D, Cai F, Belsham DD. Coordinate regulation of neuropeptide Y and agouti- related peptide gene expression by estrogen depends on the ratio of estrogen receptor (ER) alpha to ERbeta in clonal hypothalamic neurons. Mol Endocrinol 2006; 20:2080- 2092

96. Horvath TL, Bechmann I, Naftolin F, Kalra SP, Leranth C. Heterogeneity in the neuropeptide Y-containing neurons of the rat arcuate nucleus: GABAergic and non- GABAergic subpopulations. Brain research 1997; 756:283-286

97. Chaillou E, Baumont R, Chilliard Y, Tillet Y. Several subpopulations of neuropeptide Y- containing neurons exist in the infundibular nucleus of sheep: an immunohistochemical study of animals on different diets. J Comp Neurol 2002; 444:129-143

98. Dufourny L, Caraty A, Clarke IJ, Robinson JE, Skinner DC. Progesterone-receptive dopaminergic and neuropeptide Y neurons project from the arcuate nucleus to gonadotropin-releasing hormone-rich regions of the ovine preoptic area. Neuroendocrinology 2005; 82:21-31

153

99. Li C, Chen P, Smith MS. Morphological evidence for direct interaction between arcuate nucleus neuropeptide Y (NPY) neurons and gonadotropin-releasing hormone neurons and the possible involvement of NPY Y1 receptors. Endocrinology 1999; 140:5382-5390

100. Raposinho PD, Broqua P, Pierroz DD, Hayward A, Dumont Y, Quirion R, Junien JL, Aubert ML. Evidence that the inhibition of luteinizing hormone secretion exerted by central administration of neuropeptide Y (NPY) in the rat is predominantly mediated by the NPY-Y5 receptor subtype. Endocrinology 1999; 140:4046-4055

101. Barker-Gibb ML, Scott CJ, Boublik JH, Clarke IJ. The role of neuropeptide Y (NPY) in the control of LH secretion in the ewe with respect to season, NPY receptor subtype and the site of action in the hypothalamus. J Endocrinol 1995; 147:565-579

102. Hill JW, Levine JE. Abnormal response of the neuropeptide Y-deficient mouse reproductive axis to food deprivation but not lactation. Endocrinology 2003; 144:1780- 1786

103. Advis JP, Klein J, Kuljis RO, Sarkar DK, McDonald JM, Conover CA. Regulation of gonadotropin releasing hormone release by neuropeptide Y at the median eminence during the preovulatory period in ewes. Neuroendocrinology 2003; 77:246-257

104. Besecke LM, Wolfe AM, Pierce ME, Takahashi JS, Levine JE. Neuropeptide Y stimulates luteinizing hormone-releasing hormone release from superfused hypothalamic GT1-7 cells. Endocrinology 1994; 135:1621-1627

105. Bauer-Dantoin AC, McDonald JK, Levine JE. Neuropeptide Y potentiates luteinizing hormone (LH)-releasing hormone-induced LH secretion only under conditions leading to preovulatory LH surges. Endocrinology 1992; 131:2946-2952

106. Sahu A, Kalra SP. Absence of increased neuropeptide Y neuronal activity before and during the luteinizing hormone (LH) surge may underlie the attenuated preovulatory LH surge in middle-aged rats. Endocrinology 1998; 139:696-702

107. Xu M, Hill JW, Levine JE. Attenuation of luteinizing hormone surges in neuropeptide Y knockout mice. Neuroendocrinology 2000; 72:263-271

108. Kalra PS, Kalra SP. Use of antisense oligodeoxynucleotides to study the physiological functions of neuropeptide Y. Methods 2000; 22:249-254

109. Roa J, Herbison AE. Direct regulation of GnRH neuron excitability by arcuate nucleus POMC and NPY neuron neuropeptides in female mice. Endocrinology 2012; 153:5587- 5599

110. Khorram O, Pau KY, Spies HG. Bimodal effects of neuropeptide Y on hypothalamic release of gonadotropin-releasing hormone in conscious rabbits. Neuroendocrinology 1987; 45:290-297

154

111. Olofsson LE, Pierce AA, Xu AW. Functional requirement of AgRP and NPY neurons in ovarian cycle-dependent regulation of food intake. Proceedings of the National Academy of Sciences of the United States of America 2009; 106:15932-15937

112. Padilla SL, Qiu J, Nestor CC, Zhang C, Smith AW, Whiddon BB, Rønnekleiv OK, Kelly MJ, Palmiter RD. AgRP to Kiss1 neuron signaling links nutritional state and fertility. Proc Natl Acad Sci U S A 2017; 114:2413-2418

113. Santollo J, Eckel LA. Estradiol decreases the orexigenic effect of neuropeptide Y, but not agouti-related protein, in ovariectomized rats. Behavioural Brain Research 2008; 191:173-177

114. Bonavera JJ, Dube MG, Kalra PS, Kalra SP. Anorectic effects of estrogen may be mediated by decreased neuropeptide-Y release in the hypothalamic paraventricular nucleus. Endocrinology 1994; 134:2367-2370

115. Titolo D, Mayer CM, Dhillon SS, Cai F, Belsham DD. Estrogen facilitates both phosphatidylinositol 3-kinase/Akt and ERK1/2 mitogen-activated protein kinase membrane signaling required for long-term neuropeptide Y transcriptional regulation in clonal, immortalized neurons. J Neurosci 2008; 28:6473-6482

116. Mayer CM, Cai F, Cui H, Gillespie JM, MacMillan M, Belsham DD. Analysis of a repressor region in the human neuropeptide Y gene that binds Oct-1 and Pbx-1 in GT1-7 neurons. Biochem Biophys Res Commun 2003; 307:847-854

117. Dhillon SS, Belsham DD. Estrogen inhibits NPY secretion through membrane-associated estrogen receptor (ER)-alpha in clonal, immortalized hypothalamic neurons. Int J Obes (Lond) 2011; 35:198-207

118. Sar M, Sahu A, Crowley WR, Kalra SP. Localization of neuropeptide-Y immunoreactivity in estradiol-concentrating cells in the hypothalamus. Endocrinology 1990; 127:2752-2756

119. Sinchak K, Wagner EJ. Estradiol signaling in the regulation of reproduction and energy balance. Front Neuroendocrinol 2012; 33:342-363

120. Roepke TA, Qiu J, Smith AW, Rønnekleiv OK, Kelly MJ. Fasting and 17β-estradiol differentially modulate the M-current in neuropeptide Y neurons. J Neurosci 2011; 31:11825-11835

121. Korner J, Savontaus E, Chua SC, Leibel RL, Wardlaw SL. Leptin regulation of Agrp and Npy mRNA in the rat hypothalamus. J Neuroendocrinol 2001; 13:959-966

122. Hausman GJ, Barb CR, Lents CA. Leptin and reproductive function. Biochimie 2012; 94:2075-2081

123. Dhillon SS, Belsham DD. Leptin differentially regulates NPY secretion in hypothalamic cell lines through distinct intracellular signal transduction pathways. Regulatory Peptides 2011; 167:192-200

155

124. Erickson JC, Hollopeter G, Palmiter RD. Attenuation of the obesity syndrome of ob/ob mice by the loss of neuropeptide Y. Science 1996; 274:1704-1707

125. Sainsbury A, Schwarzer C, Couzens M, Jenkins A, Oakes SR, Ormandy CJ, Herzog H. Y4 receptor knockout rescues fertility in ob/ob mice. Genes Dev 2002; 16:1077-1088

126. Dhillon SS, McFadden SA, Chalmers JA, Centeno M-L, Kim GL, Belsham DD. Cellular Leptin Resistance Impairs the Leptin-Mediated Suppression of Neuropeptide Y Secretion in Hypothalamic Neurons. Endocrinology 2011; 152:4138-4147

127. Minokoshi Y, Alquier T, Furukawa N, Kim YB, Lee A, Xue B, Mu J, Foufelle F, Ferre P, Birnbaum MJ, Stuck BJ, Kahn BB. AMP-kinase regulates food intake by responding to hormonal and nutrient signals in the hypothalamus. Nature 2004; 428:569-574

128. Chen R, Wu X, Jiang L, Zhang Y. Single-Cell RNA-Seq Reveals Hypothalamic Cell Diversity. Cell Rep 2017; 18:3227-3241

129. Hanchate NK, Lee EJ, Ellis A, Kondoh K, Kuang D, Basom R, Trapnell C, Buck LB. Connect-seq to superimpose molecular on anatomical neural circuit maps. Proc Natl Acad Sci U S A 2020; 117:4375-4384

130. Kim DW, Washington PW, Wang ZQ, Lin SH, Sun C, Ismail BT, Wang H, Jiang L, Blackshaw S. The cellular and molecular landscape of hypothalamic patterning and differentiation from embryonic to late postnatal development. Nat Commun 2020; 11:4360

131. Mayer CM, Belsham DD. Insulin directly regulates NPY and AgRP gene expression via the MAPK MEK/ERK signal transduction pathway in mHypoE-46 hypothalamic neurons. Molecular and Cellular Endocrinology 2009; 307:99-108

132. Kim M-S, Pak YK, Jang P-G, Namkoong C, Choi Y-S, Won J-C, Kim K-S, Kim S-W, Kim H-S, Park J-Y, Kim Y-B, Lee K-U. Role of hypothalamic Foxo1 in the regulation of food intake and energy homeostasis. Nature Neuroscience 2006; 9:901-906

133. Kitamura T, Feng Y, Kitamura YI, Chua SC, Jr., Xu AW, Barsh GS, Rossetti L, Accili D. Forkhead protein FoxO1 mediates Agrp-dependent effects of leptin on food intake. Nat Med 2006; 12:534-540

134. Kaelin CB, Gong L, Xu AW, Yao F, Hockman K, Morton GJ, Schwartz MW, Barsh GS, MacKenzie RG. Signal transducer and activator of transcription (stat) binding sites but not stat3 are required for fasting-induced transcription of agouti-related protein messenger ribonucleic acid. Mol Endocrinol 2006; 20:2591-2602

135. Morrison CD. Leptin inhibits hypothalamic Npy and Agrp gene expression via a mechanism that requires phosphatidylinositol 3-OH-kinase signaling. AJP: Endocrinology and Metabolism 2005; 289:E1051-E1057

156

136. Tran DQ, Tse EK, Kim MH, Belsham DD. Diet-induced cellular neuroinflammation in the hypothalamus: Mechanistic insights from investigation of neurons and microglia. Molecular and Cellular Endocrinology 2016; 438:18-26

137. Fick LJ, Fick GH, Belsham DD. Palmitate alters the rhythmic expression of molecular clock genes and orexigenic neuropeptide Y mRNA levels within immortalized, hypothalamic neurons. Biochemical and biophysical research communications 2011; 413:414-419

138. Dalvi J. A.; Luo VH, D. Y.; Wellhauser, L.; Liu, Y.; Tran, D. Q.; Castel, J.; Luquet, S.; Wheeler, M. B.; Belsham, D. D., P S.; Chalmers. High fat induces acute and chronic inflammation in the hypothalamus: effect of high-fat diet, palmitate and TNF-alpha on appetite-regulating NPY neurons. Int J Obes (Lond) 2017; 41:149-158

139. Chalmers JA, Jang JJ, Belsham DD. Glucose sensing mechanisms in hypothalamic cell models: glucose inhibition of AgRP synthesis and secretion. Molecular and cellular endocrinology 2014; 382:262-270

140. Liu X, Wang N, Fan S, Zheng X, Yang Y, Zhu Y, Lu Y, Chen Q, Zhou H, Zheng J. The citrus flavonoid naringenin confers protection in a murine endotoxaemia model through AMPK-ATF3-dependent negative regulation of the TLR4 signalling pathway. Sci Rep 2016; 6:39735

141. Lee YS, Sasaki T, Kobayashi M, Kikuchi O, Kim HJ, Yokota-Hashimoto H, Shimpuku M, Susanti VY, Ido-Kitamura Y, Kimura K, Inoue H, Tanaka-Okamoto M, Ishizaki H, Miyoshi J, Ohya S, Tanaka Y, Kitajima S, Kitamura T. Hypothalamic ATF3 is involved in regulating glucose and energy metabolism in mice. Diabetologia 2013; 56:1383-1393

142. Morrison CD, Xi X, White CL, Ye J, Martin RJ. Amino acids inhibit Agrp gene expression via an mTOR-dependent mechanism. Am J Physiol Endocrinol Metab 2007; 293:E165-171

143. Lee B, Kim SG, Kim J, Choi KY, Lee S, Lee SK, Lee JW. Brain-specific homeobox factor as a target selector for glucocorticoid receptor in energy balance. Mol Cell Biol 2013; 33:2650-2658

144. Nishiyama M, Nakayama S, Iwasaki Y, Fujimoto S, Terada Y. Glucocorticoid positively regulate Agouti-related protein gene. Vol 56 P538. Endocrine Abstracts: Bioscientifica; 2018.

145. Kaelin CB, Xu AW, Lu XY, Barsh GS. Transcriptional regulation of agouti-related protein (Agrp) in transgenic mice. Endocrinology 2004; 145:5798-5806

146. Saper CB, Lu J, Chou TC, Gooley J. The hypothalamic integrator for circadian rhythms. Trends Neurosci 2005; 28:152-157

147. Oosterman JE, Kalsbeek A, la Fleur SE, Belsham DD. Impact of nutrients on circadian rhythmicity. Am J Physiol Regul Integr Comp Physiol 2015; 308:R337-350

157

148. Buijs FN, Guzman-Ruiz M, Leon-Mercado L, Basualdo MC, Escobar C, Kalsbeek A, Buijs RM. Suprachiasmatic Nucleus Interaction with the Arcuate Nucleus; Essential for Organizing Physiological Rhythms. eNeuro 2017; 4

149. Mills J, Kuohung W. Impact of circadian rhythms on female reproduction and infertility treatment success. Curr Opin Endocrinol Diabetes Obes 2019; 26:317-321

150. Marino JL, Holt VL, Chen C, Davis S. Shift work, hCLOCK T3111C polymorphism, and endometriosis risk. Epidemiology 2008; 19:477-484

151. Di Lorenzo L, De Pergola G, Zocchetti C, L'Abbate N, Basso A, Pannacciulli N, Cignarelli M, Giorgino R, Soleo L. Effect of shift work on body mass index: results of a study performed in 319 glucose-tolerant men working in a Southern Italian industry. Int J Obes Relat Metab Disord 2003; 27:1353-1358

152. Karlsson B, Knutsson A, Lindahl B. Is there an association between shift work and having a metabolic syndrome? Results from a population based study of 27,485 people. Occup Environ Med 2001; 58:747-752

153. Coomans CP, van den Berg SA, Houben T, van Klinken JB, van den Berg R, Pronk AC, Havekes LM, Romijn JA, van Dijk KW, Biermasz NR, Meijer JH. Detrimental effects of constant light exposure and high-fat diet on circadian energy metabolism and insulin sensitivity. FASEB J 2013; 27:1721-1732

154. Arble DM, Bass J, Laposky AD, Vitaterna MH, Turek FW. Circadian timing of food intake contributes to weight gain. Obesity (Silver Spring) 2009; 17:2100-2102

155. Brown-Grant K, Raisman G. Abnormalities in reproductive function associated with the destruction of the suprachiasmatic nuclei in female rats. Proc R Soc Lond B Biol Sci 1977; 198:279-296

156. Simonneaux V. A Kiss to drive rhythms in reproduction. Eur J Neurosci 2018;

157. Lehman MN, Silver R, Gladstone WR, Kahn RM, Gibson M, Bittman EL. Circadian rhythmicity restored by neural transplant. Immunocytochemical characterization of the graft and its integration with the host brain. J Neurosci 1987; 7:1626-1638

158. Herzog ED, Hermanstyne T, Smyllie NJ, Hastings MH. Regulating the Suprachiasmatic Nucleus (SCN) Circadian Clockwork: Interplay between Cell-Autonomous and Circuit- Level Mechanisms. Cold Spring Harb Perspect Biol 2017; 9

159. Zhang R, Lahens NF, Ballance HI, Hughes ME, Hogenesch JB. A circadian gene expression atlas in mammals: implications for biology and medicine. Proc Natl Acad Sci U S A 2014; 111:16219-16224

160. Turek FW, Joshu C, Kohsaka A, Lin E, Ivanova G, McDearmon E, Laposky A, Losee- Olson S, Easton A, Jensen DR, Eckel RH, Takahashi JS, Bass J. Obesity and metabolic syndrome in circadian Clock mutant mice. Science 2005; 308:1043-1045

158

161. Yang S, Liu A, Weidenhammer A, Cooksey RC, McClain D, Kim MK, Aguilera G, Abel ED, Chung JH. The role of mPer2 clock gene in glucocorticoid and feeding rhythms. Endocrinology 2009; 150:2153-2160

162. Miller BH, Olson SL, Turek FW, Levine JE, Horton TH, Takahashi JS. Circadian clock mutation disrupts estrous cyclicity and maintenance of pregnancy. Curr Biol 2004; 14:1367-1373

163. Tonsfeldt KJ, Schoeller EL, Brusman LE, Cui LJ, Lee J, Mellon PL. The Contribution of the Circadian Gene Bmal1 to Female Fertility and the Generation of the Preovulatory Luteinizing Hormone Surge. J Endocr Soc 2019; 3:716-733

164. Valenzuela FJ, Vera J, Venegas C, Munoz S, Oyarce S, Munoz K, Lagunas C. Evidences of Polymorphism Associated with Circadian System and Risk of Pathologies: A Review of the Literature. Int J Endocrinol 2016; 2016:2746909

165. Greco JA, Oosterman JE, Belsham DD. Differential effects of omega-3 fatty acid docosahexaenoic acid and palmitate on the circadian transcriptional profile of clock genes in immortalized hypothalamic neurons. Am J Physiol Regul Integr Comp Physiol 2014; 307:R1049-1060

166. Fick LJ, Fick GH, Belsham DD. Palmitate alters the rhythmic expression of molecular clock genes and orexigenic neuropeptide Y mRNA levels within immortalized, hypothalamic neurons. Biochemical and biophysical research communications 2011; 413:414-419

167. Fick LJ, Fick GH, Belsham DD. Rhythmic clock and neuropeptide gene expression in hypothalamic mHypoE-44 neurons. Mol Cell Endocrinol 2010; 323:298-306

168. Clemenzi MN, Martchenko A, Loganathan N, Tse EK, Brubaker PL, Belsham DD. Analysis of Western diet, palmitate and BMAL1 regulation of neuropeptide Y expression in the murine hypothalamus and BMAL1 knockout cell models. Mol Cell Endocrinol 2020; 507:110773

169. Johnson SA, Painter MS, Javurek AB, Ellersieck MR, Wiedmeyer CE, Thyfault JP, Rosenfeld CS. Sex-dependent effects of developmental exposure to bisphenol A and ethinyl estradiol on metabolic parameters and voluntary physical activity. Journal of developmental origins of health and disease 2015; 6:539-552

170. Monje L, Varayoud J, Muñoz-de-Toro M, Luque EH, Ramos JG. Exposure of neonatal female rats to bisphenol A disrupts hypothalamic LHRH pre-mRNA processing and estrogen receptor alpha expression in nuclei controlling estrous cyclicity. Reprod Toxicol 2010; 30:625-634

171. John N, Rehman H, Razak S, David M, Ullah W, Afsar T, Almajwal A, Alam I, Jahan S. Comparative study of environmental pollutants bisphenol A and bisphenol S on sexual differentiation of anteroventral periventricular nucleus and spermatogenesis. Reprod Biol Endocrinol 2019; 17:53

159

172. Kundakovic M, Gudsnuk K, Franks B, Madrid J, Miller RL, Perera FP, Champagne FA. Sex-specific epigenetic disruption and behavioral changes following low-dose in utero bisphenol A exposure. Proc Natl Acad Sci U S A 2013; 110:9956-9961

173. Naulé L, Picot M, Martini M, Parmentier C, Hardin-Pouzet H, Keller M, Franceschini I, Mhaouty-Kodja S. Neuroendocrine and behavioral effects of maternal exposure to oral bisphenol A in female mice. J Endocrinol 2014; 220:375-388

174. Panagiotidou E, Zerva S, Mitsiou DJ, Alexis MN, Kitraki E. Perinatal exposure to low- dose bisphenol A affects the neuroendocrine stress response in rats. J Endocrinol 2014; 220:207-218

175. Yu B, Chen QF, Liu ZP, Xu HF, Zhang XP, Xiang Q, Zhang WZ, Cui WM, Zhang X, Li N. Estrogen receptor α and β expressions in hypothalamus-pituitary-ovary axis in rats exposed lactationally to soy isoflavones and bisphenol A. Biomed Environ Sci 2010; 23:357-362

176. Castro B, Sánchez P, Torres JM, Preda O, del Moral RG, Ortega E. Bisphenol A exposure during adulthood alters expression of aromatase and 5α-reductase isozymes in rat prostate. PLoS One 2013; 8:e55905

177. Kinch CD, Ibhazehiebo K, Jeong JH, Habibi HR, Kurrasch DM. Low-dose exposure to bisphenol A and replacement bisphenol S induces precocious hypothalamic neurogenesis in embryonic zebrafish. Proc Natl Acad Sci U S A 2015; 112:1475-1480

178. Franssen D, Gérard A, Hennuy B, Donneau AF, Bourguignon JP, Parent AS. Delayed Neuroendocrine Sexual Maturation in Female Rats After a Very Low Dose of Bisphenol A Through Altered GABAergic Neurotransmission and Opposing Effects of a High Dose. Endocrinology 2016; 157:1740-1750

179. Oliveira IM, Romano RM, de Campos P, Cavallin MD, Oliveira CA, Romano MA. Delayed onset of puberty in male offspring from bisphenol A-treated dams is followed by the modulation of gene expression in the hypothalamic-pituitary-testis axis in adulthood. Reprod Fertil Dev 2017; 29:2496-2505

180. Patisaul HB, Todd KL, Mickens JA, Adewale HB. Impact of neonatal exposure to the ERalpha agonist PPT, bisphenol-A or phytoestrogens on hypothalamic kisspeptin fiber density in male and female rats. Neurotoxicology 2009; 30:350-357

181. Kurian JR, Keen KL, Kenealy BP, Garcia JP, Hedman CJ, Terasawa E. Acute Influences of Bisphenol A Exposure on Hypothalamic Release of Gonadotropin-Releasing Hormone and Kisspeptin in Female Rhesus Monkeys. Endocrinology 2015; 156:2563-2570

182. Klenke U, Constantin S, Wray S. BPA directly decreases GnRH neuronal activity via non-canonical pathway. Endocrinology 2016:en20151924-en20151924

183. MacKay H, Patterson ZR, Abizaid A. Perinatal Exposure to Low-Dose Bisphenol-A Disrupts the Structural and Functional Development of the Hypothalamic Feeding Circuitry. Endocrinology 2017; 158:768-777

160

184. Desai M, Ferrini MG, Han G, Jellyman JK, Ross MG. In vivo maternal and in vitro BPA exposure effects on hypothalamic neurogenesis and appetite regulators. Environ Res 2018; 164:45-52

185. Masuo Y, Ishido M, Morita M, Oka S. Effects of neonatal treatment with 6- hydroxydopamine and endocrine disruptors on motor activity and gene expression in rats. Neural Plast 2004; 11:59-76

186. Roepke TA, Yang JA, Yasrebi A, Mamounis KJ, Oruc E, Zama AM, Uzumcu M. Regulation of arcuate genes by developmental exposures to endocrine-disrupting compounds in female rats. Reprod Toxicol 2016; 62:18-26

187. Forner-Piquer I, Mylonas CC, Calduch-Giner J, Maradonna F, Gioacchini G, Allarà M, Piscitelli F, Di Marzo V, Pérez-Sánchez J, Carnevali O. Endocrine disruptors in the diet of male Sparus aurata: Modulation of the endocannabinoid system at the hepatic and central level by Di-isononyl phthalate and Bisphenol A. Environ Int 2018; 119:54-65

188. Rochester JR, Bolden AL. Bisphenol S and F: A Systematic Review and Comparison of the Hormonal Activity of Bisphenol A Substitutes. Environmental health perspectives 2015; 123:643-650

189. Dong S, Terasaka S, Kiyama R. Bisphenol A induces a rapid activation of Erk1/2 through GPR30 in human breast cancer cells. Environmental pollution (Barking, Essex : 1987) 2011; 159:212-218

190. Yi P, Driscoll MD, Huang J, Bhagat S, Hilf R, Bambara RA, Muyan M. The effects of estrogen-responsive element- and ligand-induced structural changes on the recruitment of cofactors and transcriptional responses by ER alpha and ER beta. Mol Endocrinol 2002; 16:674-693

191. Routledge EJ, White R, Parker MG, Sumpter JP. Differential effects of xenoestrogens on coactivator recruitment by estrogen receptor (ER) alpha and ERbeta. J Biol Chem 2000; 275:35986-35993

192. Matsushima A, Kakuta Y, Teramoto T, Koshiba T, Liu X, Okada H, Tokunaga T, Kawabata Si, Kimura M, Shimohigashi Y. Structural Evidence for Endocrine Disruptor Bisphenol A Binding to Human Nuclear Receptor ERR. Journal of Biochemistry 2007; 142:517-524

193. Takeda Y, Liu X, Sumiyoshi M, Matsushima A, Shimohigashi M, Shimohigashi Y. Placenta expressing the greatest quantity of bisphenol A receptor ERR{gamma} among the human reproductive tissues: Predominant expression of type-1 ERRgamma isoform. Journal of biochemistry 2009; 146:113-122

194. Tohmé M, Prud'homme SM, Boulahtouf A, Samarut E, Brunet F, Bernard L, Bourguet W, Gibert Y, Balaguer P, Laudet V. Estrogen-related receptor γ is an in vivo receptor of bisphenol A. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2014; 28:3124-3133

161

195. Frye C, Bo E, Calamandrei G, Calzà L, Dessì-Fulgheri F, Fernández M, Fusani L, Kah O, Kajta M, Le Page Y, Patisaul HB, Venerosi A, Wojtowicz AK, Panzica GC. Endocrine Disrupters: A Review of Some Sources, Effects, and Mechanisms of Actions on Behaviour and Neuroendocrine Systems. Journal of Neuroendocrinology 2012; 24:144- 159

196. Caserta D, Ciardo F, Bordi G, Guerranti C, Fanello E, Perra G, Borghini F, La Rocca C, Tait S, Bergamasco B, Stecca L, Marci R, Lo Monte G, Soave I, Focardi S, Mantovani A, Moscarini M. Correlation of Endocrine Disrupting Chemicals Serum Levels and White Blood Cells Gene Expression of Nuclear Receptors in a Population of Infertile Women. International Journal of Endocrinology 2013; 2013:1-7

197. Krüger T, Long M, Bonefeld-Jørgensen EC. Plastic components affect the activation of the aryl hydrocarbon and the androgen receptor. Toxicology 2008; 246:112-123

198. Ziv-Gal A, Craig ZR, Wang W, Flaws JA. Bisphenol A inhibits cultured mouse ovarian follicle growth partially via the aryl hydrocarbon receptor signaling pathway. Reproductive toxicology (Elmsford, NY) 2013; 42:58-67

199. Grygiel-Górniak B. Peroxisome proliferator-activated receptors and their ligands: nutritional and clinical implications - a review. Nutrition Journal 2014; 13:17-17

200. Garretson JT, Teubner BJW, Grove KL, Vazdarjanova A, Ryu V, Bartness TJ. Peroxisome proliferator-activated receptor γ controls ingestive behavior, agouti-related protein, and neuropeptide Y mRNA in the arcuate hypothalamus. The Journal of neuroscience : the official journal of the Society for Neuroscience 2015; 35:4571-4581

201. Dorfman MD, Thaler JP. Hypothalamic inflammation and gliosis in obesity. Current Opinion in Endocrinology & Diabetes and Obesity 2015; 22:325-330

202. Takahashi M, Komada M, Miyazawa K, Goto S, Ikeda Y. Bisphenol A exposure induces increased microglia and microglial related factors in the murine embryonic dorsal telencephalon and hypothalamus. Toxicol Lett 2018; 284:113-119

203. Salehi A, Loganathan N, Belsham DD. Bisphenol A induces Pomc gene expression through neuroinflammatory and PPARgamma nuclear receptor-mediated mechanisms in POMC-expressing hypothalamic neuronal models. Mol Cell Endocrinol 2018;

204. Huang FM, Chang YC, Lee SS, Yang ML, Kuan YH. Expression of pro-inflammatory cytokines and mediators induced by Bisphenol A via ERK-NFkappaB and JAK1/2- STAT3 pathways in macrophages. Environ Toxicol 2019; 34:486-494

205. Yang Y, Liu L, Naik I, Braunstein Z, Zhong J, Ren B. Transcription Factor C/EBP Homologous Protein in Health and Diseases. Front Immunol 2017; 8:1612

206. Ramirez S, Claret M. Hypothalamic ER stress: A bridge between leptin resistance and obesity. FEBS Lett 2015; 589:1678-1687

162

207. Asahi J, Kamo H, Baba R, Doi Y, Yamashita A, Murakami D, Hanada A, Hirano T. Bisphenol A induces endoplasmic reticulum stress-associated apoptosis in mouse non- parenchymal hepatocytes. Life Sci 2010; 87:431-438

208. Mondal A, Burchat N, Sampath H. Palmitate exacerbates bisphenol A toxicity via induction of ER stress and mitochondrial dysfunction. Biochim Biophys Acta Mol Cell Biol Lipids 2020; 1866:158816

209. Allen RG, Tresini M. Oxidative stress and gene regulation. Free Radic Biol Med 2000; 28:463-499

210. Ermak G, Davies KJ. Calcium and oxidative stress: from cell signaling to cell death. Mol Immunol 2002; 38:713-721

211. Gassman NR. Induction of oxidative stress by bisphenol A and its pleiotropic effects. Environ Mol Mutagen 2017; 58:60-71

212. Meli R, Monnolo A, Annunziata C, Pirozzi C, Ferrante MC. Oxidative Stress and BPA Toxicity: An Antioxidant Approach for Male and Female Reproductive Dysfunction. Antioxidants (Basel) 2020; 9

213. Soriano S, Ripoll C, Alonso-Magdalena P, Fuentes E, Quesada I, Nadal A, Martinez- Pinna J. Effects of Bisphenol A on ion channels: Experimental evidence and molecular mechanisms. Steroids 2016; 111:12-20

214. Zhong X, Li J, Zhuang Z, Shen Q, Jiang K, Hu Y, Wu D, Xu X. Rapid effect of bisphenol A on glutamate-induced Ca. Mol Cell Endocrinol 2019; 485:35-43

215. Wang H, Zhao P, Huang Q, Chi Y, Dong S, Fan J. Bisphenol-A induces neurodegeneration through disturbance of intracellular calcium homeostasis in human embryonic stem cells-derived cortical neurons. Chemosphere 2019; 229:618-630

216. Yang YJ, Hong YC, Oh SY, Park MS, Kim H, Leem JH, Ha EH. Bisphenol A exposure is associated with oxidative stress and inflammation in postmenopausal women. Environ Res 2009; 109:797-801

217. Hinchy EC, Gruszczyk AV, Willows R, Navaratnam N, Hall AR, Bates G, Bright TP, Krieg T, Carling D, Murphy MP. Mitochondria-derived ROS activate AMP-activated protein kinase (AMPK) indirectly. J Biol Chem 2018; 293:17208-17217

218. Cavaliere G, Viggiano E, Trinchese G, De Filippo C, Messina A, Monda V, Valenzano A, Cincione RI, Zammit C, Cimmino F, Catapano A, Sessa F, Messina G, Monda M, Crispino M, Mollica MP. Long Feeding High-Fat Diet Induces Hypothalamic Oxidative Stress and Inflammation, and Prolonged Hypothalamic AMPK Activation in Rat Animal Model. Front Physiol 2018; 9:818

219. Belsham DD, Cai F, Cui H, Smukler SR, Salapatek AMF, Shkreta L. Generation of a phenotypic array of hypothalamic neuronal cell models to study complex neuroendocrine disorders. Endocrinology 2004; 145:393-400

163

220. Belsham DD, Fick LJ, Dalvi PS, Centeno ML, Chalmers JA, Lee PK, Wang Y, Drucker DJ, Koletar MM. Ciliary neurotrophic factor recruitment of -like peptide-1 mediates neurogenesis, allowing immortalization of adult murine hypothalamic neurons. FASEB J 2009; 23:4256-4265

221. Treen AK, Luo V, Chalmers JA, Dalvi PS, Tran D, Ye W, Kim GL, Friedman Z, Belsham DD. Divergent Regulation of ER and Kiss Genes by 17β-Estradiol in Hypothalamic ARC Versus AVPV Models. Molecular Endocrinology 2016; 30:217-233

222. Ogo Y, Taniuchi S, Ojima F, Hayashi S, Murakami I, Saito Y, Takeuchi S, Kudo T, Takahashi S. IGF-1 gene expression is differentially regulated by estrogen receptors alpha and beta in mouse endometrial stromal cells and ovarian granulosa cells. J Reprod Dev 2014; 60:216-223

223. Lee JH, Kim EJ, Kim DK, Lee JM, Park SB, Lee IK, Harris RA, Lee MO, Choi HS. Hypoxia induces PDK4 gene expression through induction of the orphan nuclear receptor ERRgamma. PLoS One 2012; 7:e46324

224. Revest JM, Di Blasi F, Kitchener P, Rouge-Pont F, Desmedt A, Turiault M, Tronche F, Piazza PV. The MAPK pathway and Egr-1 mediate stress-related behavioral effects of glucocorticoids. Nat Neurosci 2005; 8:664-672

225. Loganathan N, Salehi A, Chalmers JA, Belsham DD. Bisphenol A Alters Bmal1, Per2, and Rev-Erba mRNA and Requires Bmal1 to Increase Neuropeptide Y Expression in Hypothalamic Neurons. Endocrinology 2019; 160:181-192

226. Zhao J, Li X, Guo M, Yu J, Yan C. The common stress responsive transcription factor ATF3 binds genomic sites enriched with p300 and H3K27ac for transcriptional regulation. BMC Genomics 2016; 17:335

227. Marraudino M, Bonaldo B, Farinetti A, Panzica G, Ponti G, Gotti S. Metabolism Disrupting Chemicals and Alteration of Neuroendocrine Circuits Controlling Food Intake and Energy Metabolism. Front Endocrinol (Lausanne) 2018; 9:766

228. Stats C. Bisphenol A concentrations in Canadians, 2012 and 2013. 2013.

229. Provvisiero DP, Pivonello C, Muscogiuri G, Negri M, de Angelis C, Simeoli C, Pivonello R, Colao A. Influence of Bisphenol A on Type 2 Diabetes Mellitus. Int J Environ Res Public Health 2016; 13

230. Iurlaro R, Munoz-Pinedo C. Cell death induced by endoplasmic reticulum stress. FEBS J 2016; 283:2640-2652

231. Park S, Jang A, Bouret SG. Maternal obesity-induced endoplasmic reticulum stress causes metabolic alterations and abnormal hypothalamic development in the offspring. PLoS Biol 2020; 18:e3000296

164

232. Calafat AM, Kuklenyik Z, Reidy JA, Caudill SP, Ekong J, Needham LL. Urinary concentrations of bisphenol A and 4-nonylphenol in a human reference population. Environ Health Perspect 2005; 113:391-395

233. Jin-Hua Li B-XZ, Wei-Min Cai. The Solubility Behavior of Bisphenol A in the Presence of Surfactants. J Chem Eng Data 2007; 52:2511-2513

234. Mayer SI, Dexheimer V, Nishida E, Kitajima S, Thiel G. Expression of the transcriptional repressor ATF3 in gonadotrophs is regulated by Egr-1, CREB, and ATF2 after gonadotropin-releasing hormone receptor stimulation. Endocrinology 2008; 149:6311-6325

235. Mungrue IN, Pagnon J, Kohannim O, Gargalovic PS, Lusis AJ. CHAC1/MGC4504 is a novel proapoptotic component of the unfolded protein response, downstream of the ATF4-ATF3-CHOP cascade. J Immunol 2009; 182:466-476

236. Lu D, Chen J, Hai T. The regulation of ATF3 gene expression by mitogen-activated protein kinases. Biochem J 2007; 401:559-567

237. Tran A, Loganathan N, McIlwraith EK, Belsham DD. Palmitate and Nitric Oxide Regulate the Expression of Spexin and Receptors 2 and 3 in Hypothalamic Neurons. Neuroscience 2019;

238. Wei J, Lin Y, Li Y, Ying C, Chen J, Song L, Zhou Z, Lv Z, Xia W, Chen X, Xu S. Perinatal Exposure to Bisphenol A at Reference Dose Predisposes Offspring to Metabolic Syndrome in Adult Rats on a High-Fat Diet. Endocrinology 2011;

239. Ropero AB, Alonso-Magdalena P, Garcia-Garcia E, Ripoll C, Fuentes E, Nadal A. Bisphenol-A disruption of the endocrine pancreas and blood glucose homeostasis. Int J Androl 2008; 31:194-200

240. Marmugi A, Ducheix S, Lasserre F, Polizzi A, Paris A, Priymenko N, Bertrand-Michel J, Pineau T, Guillou H, Martin PGP, Mselli-Lakhal L. Low doses of bisphenol A induce gene expression related to lipid synthesis and trigger triglyceride accumulation in adult mouse liver. Hepatology (Baltimore, Md) 2012; 55:395-407

241. Acconcia F, Pallottini V, Marino M. Molecular Mechanisms of Action of BPA. Dose Response 2015; 13:1559325815610582

242. Otto C, Rohde-Schulz B, Schwarz G, Fuchs I, Klewer M, Brittain D, Langer G, Bader B, Prelle K, Nubbemeyer R, Fritzemeier KH. G protein-coupled receptor 30 localizes to the endoplasmic reticulum and is not activated by estradiol. Endocrinology 2008; 149:4846- 4856

243. Revankar CM, Cimino DF, Sklar LA, Arterburn JB, Prossnitz ER. A transmembrane intracellular estrogen receptor mediates rapid cell signaling. Science 2005; 307:1625- 1630

165

244. Revankar CM, Mitchell HD, Field AS, Burai R, Corona C, Ramesh C, Sklar LA, Arterburn JB, Prossnitz ER. Synthetic estrogen derivatives demonstrate the functionality of intracellular GPR30. ACS Chem Biol 2007; 2:536-544

245. Funakoshi T, Yanai A, Shinoda K, Kawano MM, Mizukami Y. G protein-coupled receptor 30 is an estrogen receptor in the plasma membrane. Biochem Biophys Res Commun 2006; 346:904-910

246. Giraldo A, Barrett OP, Tindall MJ, Fuller SJ, Amirak E, Bhattacharya BS, Sugden PH, Clerk A. Feedback regulation by Atf3 in the -1-responsive transcriptome of cardiomyocytes: Egr1 is a principal Atf3 target. Biochem J 2012; 444:343-355

247. Riedel CS, Georg B, Jorgensen HL, Hannibal J, Fahrenkrug J. Mice Lacking EGR1 Have Impaired Clock Gene (BMAL1) Oscillation, Locomotor Activity, and Body Temperature. J Mol Neurosci 2018; 64:9-19

248. Allison MB, Pan W, MacKenzie A, Patterson C, Shah K, Barnes T, Cheng W, Rupp A, Olson DP, Myers MG, Jr. Defining the Transcriptional Targets of Leptin Reveals a Role for Atf3 in Leptin Action. Diabetes 2018; 67:1093-1104

249. Valentino R, D'Esposito V, Passaretti F, Liotti A, Cabaro S, Longo M, Perruolo G, Oriente F, Beguinot F, Formisano P. Bisphenol-A impairs insulin action and up-regulates inflammatory pathways in human subcutaneous adipocytes and 3T3-L1 cells. PLoS One 2013; 8:e82099

250. Varela L, Horvath TL. Leptin and insulin pathways in POMC and AgRP neurons that modulate energy balance and glucose homeostasis. EMBO Rep 2012; 13:1079-1086

251. Mesaros A, Koralov SB, Rother E, Wunderlich FT, Ernst MB, Barsh GS, Rajewsky K, Bruning JC. Activation of Stat3 signaling in AgRP neurons promotes locomotor activity. Cell Metab 2008; 7:236-248

252. Prasanth GK, Divya LM, Sadasivan C. Bisphenol-A can bind to human glucocorticoid receptor as an agonist: an in silico study. J Appl Toxicol 2010; 30:769-774

253. Sakkou M, Wiedmer P, Anlag K, Hamm A, Seuntjens E, Ettwiller L, Tschop MH, Treier M. A role for brain-specific homeobox factor Bsx in the control of hyperphagia and locomotory behavior. Cell Metab 2007; 5:450-463

254. Ilnytska O, Stutz AM, Park-York M, York DA, Ribnicky DM, Zuberi A, Cefalu WT, Argyropoulos G. Molecular mechanisms for activation of the agouti-related protein and stimulation of appetite. Diabetes 2011; 60:97-106

255. Whitlock NC, Bahn JH, Lee SH, Eling TE, Baek SJ. -induced apoptosis is mediated by early growth response-1, Kruppel-like factor 4, and activating transcription factor 3. Cancer Prev Res (Phila) 2011; 4:116-127

166

256. Ilnytska O, Sozen MA, Dauterive R, Argyropoulos G. Control elements in the neighboring ATPase gene influence spatiotemporal expression of the human agouti- related protein. J Mol Biol 2009; 388:239-251

257. Shaul O. How introns enhance gene expression. Int J Biochem Cell Biol 2017; 91:145- 155

258. Marklund U, Bystrom M, Gedda K, Larefalk A, Juneblad K, Nystrom S, Ekstrand AJ. Intron-mediated expression of the human neuropeptide Y Y1 receptor. Mol Cell Endocrinol 2002; 188:85-97

259. Pan YX, Chen H, Thiaville MM, Kilberg MS. Activation of the ATF3 gene through a co- ordinated amino acid-sensing response programme that controls transcriptional regulation of responsive genes following amino acid limitation. Biochem J 2007; 401:299-307

260. Lage R, Vazquez MJ, Varela L, Saha AK, Vidal-Puig A, Nogueiras R, Dieguez C, Lopez M. Ghrelin effects on neuropeptides in the rat hypothalamus depend on fatty acid metabolism actions on BSX but not on gender. FASEB J 2010; 24:2670-2679

261. Sunaga H, Matsui H, Anjo S, Syamsunarno MR, Koitabashi N, Iso T, Matsuzaka T, Shimano H, Yokoyama T, Kurabayashi M. Elongation of Long-Chain Fatty Acid Family Member 6 (Elovl6)-Driven Fatty Acid Metabolism Regulates Vascular Smooth Muscle Cell Phenotype Through AMP-Activated Protein Kinase/Kruppel-Like Factor 4 (AMPK/KLF4) Signaling. J Am Heart Assoc 2016; 5

262. Ruud J, Steculorum SM, Bruning JC. Neuronal control of peripheral insulin sensitivity and glucose metabolism. Nat Commun 2017; 8:15259

263. Fang F, Chen D, Yu P, Qian W, Zhou J, Liu J, Gao R, Wang J, Xiao H. Effects of Bisphenol A on glucose homeostasis and brain insulin signaling pathways in male mice. General and comparative endocrinology 2015; 212:44-50

264. Stanley BG, Kyrkouli SE, Lampert S, Leibowitz SF. Neuropeptide Y chronically injected into the hypothalamus: a powerful neurochemical inducer of hyperphagia and obesity. Peptides 7:1189-1192

265. Canada S. Bisphenol A concentrations in Canadians, 2012 and 2013. 2013;

266. Shen Y, Zheng Y, Jiang J, Liu Y, Luo X, Shen Z, Chen X, Wang Y, Dai Y, Zhao J, Liang H, Chen A, Yuan W. Higher urinary bisphenol A concentration is associated with unexplained recurrent miscarriage risk: evidence from a case-control study in eastern China. PLoS One 2015; 10:e0127886

267. Wisniewski P, Romano RM, Kizys MM, Oliveira KC, Kasamatsu T, Giannocco G, Chiamolera MI, Dias-da-Silva MR, Romano MA. Adult exposure to bisphenol A (BPA) in Wistar rats reduces sperm quality with disruption of the hypothalamic-pituitary- testicular axis. Toxicology 2015; 329:1-9

167

268. Al-Hiyasat AS, Darmani H, Elbetieha AM. Leached components from dental composites and their effects on fertility of female mice. Eur J Oral Sci 2004; 112:267-272

269. Loganathan N, McIlwraith EK, Belsham DD. Bisphenol A induces Agrp gene expression in hypothalamic neurons through a mechanism involving ATF3. Neuroendocrinology 2020;

270. MacKay H, Abizaid A. A plurality of molecular targets: The receptor ecosystem for bisphenol-A (BPA). Horm Behav 2018; 101:59-67

271. Vomund S, Schafer A, Parnham MJ, Brune B, von Knethen A. Nrf2, the Master Regulator of Anti-Oxidative Responses. Int J Mol Sci 2017; 18

272. Brewer GJ, Torricelli JR, Evege EK, Price PJ. Optimized survival of hippocampal neurons in B27-supplemented Neurobasal, a new serum-free medium combination. J Neurosci Res 1993; 35:567-576

273. Castro-Caldas M, Carvalho AN, Rodrigues E, Henderson CJ, Wolf CR, Rodrigues CM, Gama MJ. Tauroursodeoxycholic acid prevents MPTP-induced dopaminergic cell death in a mouse model of Parkinson's disease. Mol Neurobiol 2012; 46:475-486

274. Rosa AI, Fonseca I, Nunes MJ, Moreira S, Rodrigues E, Carvalho AN, Rodrigues CMP, Gama MJ, Castro-Caldas M. Novel insights into the antioxidant role of tauroursodeoxycholic acid in experimental models of Parkinson's disease. Biochim Biophys Acta Mol Basis Dis 2017; 1863:2171-2181

275. Floriano-Sanchez E, Villanueva C, Medina-Campos ON, Rocha D, Sanchez-Gonzalez DJ, Cardenas-Rodriguez N, Pedraza-Chaverri J. Nordihydroguaiaretic acid is a potent in vitro scavenger of peroxynitrite, singlet oxygen, hydroxyl radical, superoxide anion and hypochlorous acid and prevents in vivo ozone-induced tyrosine nitration in lungs. Free Radic Res 2006; 40:523-533

276. Kim SY, Kim TB, Moon KA, Kim TJ, Shin D, Cho YS, Moon HB, Lee KY. Regulation of pro-inflammatory responses by lipoxygenases via intracellular reactive oxygen species in vitro and in vivo. Exp Mol Med 2008; 40:461-476

277. Sahu A, Kalra PS, Crowley WR, Kalra SP. Functional heterogeneity in neuropeptide-Y- producing cells in the rat brain as revealed by testosterone action. Endocrinology 1990; 127:2307-2312

278. Kim GL, Dhillon SS, Belsham DD. Kisspeptin directly regulates neuropeptide Y synthesis and secretion via the ERK1/2 and p38 mitogen-activated protein kinase signaling pathways in NPY-secreting hypothalamic neurons. Endocrinology 2010; 151:5038-5047

279. Bulzomi P, Bolli A, Marino M. Estrogen receptor-dependent effects of bisphenol a. J Biol Res-Boll Soc 2011; 84:91-94

168

280. Xu XB, He Y, Song C, Ke X, Fan SJ, Peng WJ, Tan R, Kawata M, Matsuda K, Pan BX, Kato N. Bisphenol A regulates the estrogen receptor alpha signaling in developing hippocampus of male rats through estrogen receptor. Hippocampus 2014; 24:1570-1580

281. Cao J, Joyner L, Mickens JA, Leyrer SM, Patisaul HB. Sex-specific Esr2 mRNA expression in the rat hypothalamus and amygdala is altered by neonatal bisphenol A exposure. Reproduction 2014; 147:537-554

282. Frank A, Brown LM, Clegg DJ. The role of hypothalamic estrogen receptors in metabolic regulation. Frontiers in Neuroendocrinology 2014; 35:550-557

283. Drougard A, Fournel A, Valet P, Knauf C. Impact of hypothalamic reactive oxygen species in the regulation of energy metabolism and food intake. Front Neurosci 2015; 9:56

284. Numakawa Y, Matsumoto T, Yokomaku D, Taguchi T, Niki E, Hatanaka H, Kunugi H, Numakawa T. 17beta-estradiol protects cortical neurons against oxidative stress-induced cell death through reduction in the activity of mitogen-activated protein kinase and in the accumulation of intracellular calcium. Endocrinology 2007; 148:627-637

285. Hsu CC, Cheng CH, Hsu CL, Lee WJ, Huang SC, Huang YC. Role of vitamin B6 status on antioxidant defenses, glutathione, and related enzyme activities in mice with homocysteine-induced oxidative stress. Food Nutr Res 2015; 59:25702

286. Aldini G, Altomare A, Baron G, Vistoli G, Carini M, Borsani L, Sergio F. N- Acetylcysteine as an antioxidant and disulphide breaking agent: the reasons why. Free Radic Res 2018; 52:751-762

287. Irrcher I, Ljubicic V, Hood DA. Interactions between ROS and AMP kinase activity in the regulation of PGC-1alpha transcription in skeletal muscle cells. Am J Physiol Cell Physiol 2009; 296:C116-123

288. Susiarjo M, Xin F, Stefaniak M, Mesaros C, Simmons RA, Bartolomei MS. Bile Acids and Tryptophan Metabolism Are Novel Pathways Involved in Metabolic Abnormalities in BPA-Exposed Pregnant Mice and Male Offspring. Endocrinology 2017; 158:2533- 2542

289. Rodriguez-Rodriguez E, Lopez-Sobaler AM, Navarro AR, Bermejo LM, Ortega RM, Andres P. Vitamin B6 status improves in overweight/obese women following a hypocaloric diet rich in breakfast cereals, and may help in maintaining fat-free mass. Int J Obes (Lond) 2008; 32:1552-1558

290. Zemel MB, Bruckbauer A. Effects of a leucine and pyridoxine-containing nutraceutical on fat oxidation, and oxidative and inflammatory stress in overweight and obese subjects. Nutrients 2012; 4:529-541

291. Grosser N, Oberle S, Berndt G, Erdmann K, Hemmerle A, Schroder H. Antioxidant action of L-alanine: heme oxygenase-1 and ferritin as possible mediators. Biochem Biophys Res Commun 2004; 314:351-355

169

292. Wang X, Perez E, Liu R, Yan LJ, Mallet RT, Yang SH. Pyruvate protects mitochondria from oxidative stress in human neuroblastoma SK-N-SH cells. Brain Res 2007; 1132:1-9

293. Lee SR. Critical Role of Zinc as Either an Antioxidant or a Prooxidant in Cellular Systems. Oxid Med Cell Longev 2018; 2018:9156285

294. Dhillon SS, Gingerich S, Virtanen C, Belsham DD. Gene array analysis of embryonic- versus adult-derived hypothalamic NPY-expressing cell lines. Molecular and cellular endocrinology 2012; 358:116-126

295. Friedman MJ, Li S, Li XJ. Activation of gene transcription by heat shock protein 27 may contribute to its neuronal protection. J Biol Chem 2009; 284:27944-27951

296. Miller H, Poon S, Hibbert B, Rayner K, Chen YX, O'Brien ER. Modulation of estrogen signaling by the novel interaction of heat shock protein 27, a biomarker for atherosclerosis, and estrogen receptor beta: mechanistic insight into the vascular effects of . Arterioscler Thromb Vasc Biol 2005; 25:e10-14

297. Andersson G, Pahlman S, Parrow V, Johansson I, Hammerling U. Activation of the human NPY gene during neuroblastoma cell differentiation: induced transcriptional activities of AP-1 and AP-2. Cell Growth Differ 1994; 5:27-36

298. De Souza Santos R, Frank AP, Palmer BF, Clegg DJ. Sex and media: Considerations for cell culture studies. ALTEX 2018; 35:435-440

299. Wellhauser L, Gojska NM, Belsham DD. Delineating the regulation of energy homeostasis using hypothalamic cell models. Frontiers in Neuroendocrinology 2015; 36:130-149

300. Dalvi A; Tung SB, D.D, P S.; Nazarians-Armavil. Immortalized neurons for the study of hypothalamic function. Am J Physiol Regul Integr Comp Physiol 2011; 300:23

301. Oosterman JE, Kalsbeek A, la Fleur SE, Belsham DD. Impact of nutrients on circadian rhythmicity. American journal of physiology Regulatory, integrative and comparative physiology 2015; 308:R337-R350

302. Nagai K, Nishio T, Nakagawa H, Nakamura S, Fukuda Y. Effect of bilateral lesions of the suprachiasmatic nuclei on the circadian rhythm of food-intake. Brain Res 1978; 142:384-389

303. Kohsaka A, Laposky AD, Ramsey KM, Estrada C, Joshu C, Kobayashi Y, Turek FW, Bass J. High-fat diet disrupts behavioral and molecular circadian rhythms in mice. Cell Metab 2007; 6:414-421

304. Alonso-Magdalena P, Quesada I, Nadal A. Prenatal Exposure to BPA and Offspring Outcomes: The Diabesogenic Behavior of BPA. Dose Response 2015; 13:1559325815590395

170

305. Yang M, Chen M, Wang J, Xu M, Sun J, Ding L, Lv X, Ma Q, Bi Y, Liu R, Hong J, Ning G. Bisphenol A Promotes Adiposity and Inflammation in a Nonmonotonic Dose-response Way in 5-week-old Male and Female C57BL/6J Mice Fed a Low-calorie Diet. Endocrinology 2016; 157:2333-2345

306. Chalmers JA, Lin SY, Martino TA, Arab S, Liu P, Husain M, Sole MJ, Belsham DD. Diurnal profiling of neuroendocrine genes in murine heart, and shift in gene expression with pressure-overload cardiac hypertrophy. J Mol Endocrinol 2008; 41:117-124

307. Stutz AM, Staszkiewicz J, Ptitsyn A, Argyropoulos G. Circadian expression of genes regulating food intake. Obesity (Silver Spring) 2007; 15:607-615

308. Preitner N, Damiola F, Lopez-Molina L, Zakany J, Duboule D, Albrecht U, Schibler U. The orphan nuclear receptor REV-ERBalpha controls circadian transcription within the positive limb of the mammalian circadian oscillator. Cell 2002; 110:251-260

309. Panda S, Antoch MP, Miller BH, Su AI, Schook AB, Straume M, Schultz PG, Kay SA, Takahashi JS, Hogenesch JB. Coordinated transcription of key pathways in the mouse by the circadian clock. Cell 2002; 109:307-320

310. Lowrey PL, Takahashi JS. Mammalian circadian biology: elucidating genome-wide levels of temporal organization. Annu Rev Genomics Hum Genet 2004; 5:407-441

311. Boden G. Interaction between free fatty acids and glucose metabolism. Curr Opin Clin Nutr Metab Care 2002; 5:545-549

312. Angelousi A, Kassi E, Nasiri-Ansari N, Weickert MO, Randeva H, Kaltsas G. Clock genes alterations and endocrine disorders. Eur J Clin Invest 2018; 48:e12927

313. Chevalier N, Fenichel P. Bisphenol A: Targeting metabolic tissues. Rev Endocr Metab Disord 2015; 16:299-309

314. Hao H, Allen DL, Hardin PE. A circadian enhancer mediates PER-dependent mRNA cycling in Drosophila melanogaster. Mol Cell Biol 1997; 17:3687-3693

315. Yoshitane H, Ozaki H, Terajima H, Du NH, Suzuki Y, Fujimori T, Kosaka N, Shimba S, Sugano S, Takagi T, Iwasaki W, Fukada Y. CLOCK-controlled polyphonic regulation of circadian rhythms through canonical and noncanonical E-boxes. Mol Cell Biol 2014; 34:1776-1787

316. Ripperger JA, Shearman LP, Reppert SM, Schibler U. CLOCK, an essential pacemaker component, controls expression of the circadian transcription factor DBP. Genes Dev 2000; 14:679-689

317. Oosterman JE, Belsham DD. Glucose Alters Per2 Rhythmicity Independent of AMPK, Whereas AMPK Inhibitor Compound C Causes Profound Repression of Clock Genes and AgRP in mHypoE-37 Hypothalamic Neurons. PLoS One 2016; 11:e0146969

171

318. Nesan D, Antle M, Kurrasch D. Low-Dose Gestational BPA Exposure Alters Circadian Rhythms in Mice. 2019; 3

319. Weber DN, Hoffmann RG, Hoke ES, Tanguay RL. Bisphenol A exposure during early development induces sex-specific changes in adult zebrafish social interactions. J Toxicol Environ Health A 2015; 78:50-66

320. Sen A, Sellix MT. The Circadian Timing System and Environmental Circadian Disruption: From Follicles to Fertility. Endocrinology 2016; 157:3366-3373

321. Choi JY, Choe JR, Lee TH, Choi CY. Effects of bisphenol A and light conditions on the circadian rhythm of the goldfish Carassius auratus. Biological Rhythm Research 2018; 49:502-514

322. Walker DM, Goetz BM, Gore AC. Dynamic postnatal developmental and sex-specific neuroendocrine effects of prenatal polychlorinated biphenyls in rats. Mol Endocrinol 2014; 28:99-115

323. Ali AAH, Schwarz-Herzke B, Mir S, Sahlender B, Victor M, Gorg B, Schmuck M, Dach K, Fritsche E, Kremer A, von Gall C. Deficiency of the clock gene Bmal1 affects neural progenitor cell migration. Brain Struct Funct 2018;

324. Rudic RD, McNamara P, Curtis AM, Boston RC, Panda S, Hogenesch JB, Fitzgerald GA. BMAL1 and CLOCK, two essential components of the circadian clock, are involved in glucose homeostasis. PLoS Biol 2004; 2:e377

325. Shi S, Hida A, McGuinness OP, Wasserman DH, Yamazaki S, Johnson CH. Circadian clock gene Bmal1 is not essential; functional replacement with its paralog, Bmal2. Curr Biol 2010; 20:316-321

326. Birky TL, Bray MS. Understanding circadian gene function: animal models of tissue- specific circadian disruption. IUBMB Life 2014; 66:34-41

327. Orozco-Solis R, Matos RJ, Lopes de Souza S, Grit I, Kaeffer B, Manhaes de Castro R, Bolanos-Jimenez F. Perinatal nutrient restriction induces long-lasting alterations in the circadian expression pattern of genes regulating food intake and energy metabolism. Int J Obes (Lond) 2011; 35:990-1000

328. Mieda M, Sakurai T. Bmal1 in the nervous system is essential for normal adaptation of circadian locomotor activity and food intake to periodic feeding. J Neurosci 2011; 31:15391-15396

329. Dumbell R, Leliavski A, Matveeva O, Blaum C, Tsang AH, Oster H. Dissociation of Molecular and Endocrine Circadian Rhythms in Male Mice Lacking Bmal1 in the Adrenal Cortex. Endocrinology 2016; 157:4222-4233

330. Acconcia F, Pallottini V, Marino M. Molecular Mechanisms of Action of BPA. Dose- Response 2015; 13:1559325815610582

172

331. Rhee JS, Kim BM, Lee BY, Hwang UK, Lee YS, Lee JS. Cloning of circadian rhythmic pathway genes and perturbation of oscillation patterns in endocrine disrupting chemicals (EDCs)-exposed mangrove killifish Kryptolebias marmoratus. Comp Biochem Physiol C Toxicol Pharmacol 2014; 164:11-20

332. Zhu J, Jiang L, Liu Y, Qian W, Liu J, Zhou J, Gao R, Xiao H, Wang J. MAPK and NF- kappaB pathways are involved in bisphenol A-induced TNF-alpha and IL-6 production in BV2 microglial cells. Inflammation 2015; 38:637-648

333. Amato S, Man HY. Bioenergy sensing in the brain: the role of AMP-activated protein kinase in neuronal metabolism, development and neurological diseases. Cell Cycle 2011; 10:3452-3460

334. Ramamurthy S, Ronnett G. AMP-Activated Protein Kinase (AMPK) and Energy-Sensing in the Brain. Exp Neurobiol 2012; 21:52-60

335. Turnley AM, Stapleton D, Mann RJ, Witters LA, Kemp BE, Bartlett PF. Cellular distribution and developmental expression of AMP-activated protein kinase isoforms in mouse central nervous system. J Neurochem 1999; 72:1707-1716

336. Hawley SA, Pan DA, Mustard KJ, Ross L, Bain J, Edelman AM, Frenguelli BG, Hardie DG. Calmodulin-dependent protein kinase kinase-beta is an alternative upstream kinase for AMP-activated protein kinase. Cell Metab 2005; 2:9-19

337. Dzamko N, van Denderen BJ, Hevener AL, Jørgensen SB, Honeyman J, Galic S, Chen ZP, Watt MJ, Campbell DJ, Steinberg GR, Kemp BE. AMPK beta1 deletion reduces appetite, preventing obesity and hepatic insulin resistance. J Biol Chem 2010; 285:115- 122

338. Kjøbsted R, Hingst JR, Fentz J, Foretz M, Sanz MN, Pehmøller C, Shum M, Marette A, Mounier R, Treebak JT, Wojtaszewski JFP, Viollet B, Lantier L. AMPK in skeletal muscle function and metabolism. FASEB J 2018; 32:1741-1777

339. Steinberg GR, O'Neill HM, Dzamko NL, Galic S, Naim T, Koopman R, Jørgensen SB, Honeyman J, Hewitt K, Chen ZP, Schertzer JD, Scott JW, Koentgen F, Lynch GS, Watt MJ, van Denderen BJ, Campbell DJ, Kemp BE. Whole body deletion of AMP-activated protein kinase {beta}2 reduces muscle AMPK activity and exercise capacity. J Biol Chem 2010; 285:37198-37209

340. Jordan SD, Lamia KA. AMPK at the crossroads of circadian clocks and metabolism. Mol Cell Endocrinol 2013; 366:163-169

341. Kim SM, Neuendorff N, Chapkin RS, Earnest DJ. Role of Inflammatory Signaling in the Differential Effects of Saturated and Poly-unsaturated Fatty Acids on Peripheral Circadian Clocks. EBioMedicine 2016; 7:100-111

342. Harfmann BD, Schroder EA, Kachman MT, Hodge BA, Zhang X, Esser KA. Muscle- specific loss of Bmal1 leads to disrupted tissue glucose metabolism and systemic glucose homeostasis. Skelet Muscle 2016; 6:12

173

343. Goncalves R, Zanatta AP, Cavalari FC, do Nascimento MAW, Delalande-Lecapitaine C, Bouraima-Lelong H, Silva F. Acute effect of bisphenol A: Signaling pathways on calcium influx in immature rat testes. Reprod Toxicol 2018; 77:94-102

344. Deutschmann A, Hans M, Meyer R, Haberlein H, Swandulla D. Bisphenol A inhibits voltage-activated Ca(2+) channels in vitro: mechanisms and structural requirements. Mol Pharmacol 2013; 83:501-511

345. Cai W, Rambaud J, Teboul M, Masse I, Benoit G, Gustafsson JA, Delaunay F, Laudet V, Pongratz I. Expression levels of estrogen receptor beta are modulated by components of the molecular clock. Mol Cell Biol 2008; 28:784-793

346. McMinn JE, Seeley RJ, Wilkinson CW, Havel PJ, Woods SC, Schwartz MW. NPY- induced overfeeding suppresses hypothalamic NPY mRNA expression: potential roles of plasma insulin and leptin. Regul Pept 1998; 75-76:425-431

347. Zarjevski N, Cusin I, Vettor R, Rohner-Jeanrenaud F, Jeanrenaud B. Intracerebroventricular administration of neuropeptide Y to normal rats has divergent effects on glucose utilization by adipose tissue and skeletal muscle. Diabetes 1994; 43:764-769

348. Goto K, Inui A, Takimoto Y, Yuzuriha H, Asakawa A, Kawamura Y, Tsuji H, Takahara Y, Takeyama C, Katsuura G, Kasuga M. Acute intracerebroventricular administration of either carboxyl-terminal or amino-terminal fragments of agouti-related peptide produces a long-term decrease in energy expenditure in rats. Int J Mol Med 2003; 12:379-383

349. Ip CK, Zhang L, Farzi A, Qi Y, Clarke I, Reed F, Shi YC, Enriquez R, Dayas C, Graham B, Begg D, Brüning JC, Lee NJ, Hernandez-Sanchez D, Gopalasingam G, Koller J, Tasan R, Sperk G, Herzog H. Amygdala NPY Circuits Promote the Development of Accelerated Obesity under Chronic Stress Conditions. Cell Metab 2019; 30:111-128.e116

350. Goto M, Arima H, Watanabe M, Hayashi M, Banno R, Sato I, Nagasaki H, Oiso Y. Ghrelin increases neuropeptide Y and agouti-related peptide gene expression in the arcuate nucleus in rat hypothalamic organotypic cultures. Endocrinology 2006; 147:5102- 5109

351. Nieminen P, Lindstrom-Seppa P, Mustonen AM, Mussalo-Rauhamaa H, Kukkonen JV. Bisphenol A affects endocrine physiology and biotransformation enzyme activities of the field vole (Microtus agrestis). Gen Comp Endocrinol 2002; 126:183-189

352. Ronn M, Lind L, Orberg J, Kullberg J, Soderberg S, Larsson A, Johansson L, Ahlstrom H, Lind PM. Bisphenol A is related to circulating levels of adiponectin, leptin and ghrelin, but not to fat mass or fat distribution in humans. Chemosphere 2014; 112:42-48

353. Nakazato M, Murakami N, Date Y, Kojima M, Matsuo H, Kangawa K, Matsukura S. A role for ghrelin in the central regulation of feeding. Nature 2001; 409:194-198

354. Mustieles V, Ocon-Hernandez O, Minguez-Alarcon L, Davila-Arias C, Perez-Lobato R, Calvente I, Arrebola JP, Vela-Soria F, Rubio S, Hauser R, Olea N, Fernandez MF.

174

Bisphenol A and reproductive hormones and cortisol in peripubertal boys: The INMA- Granada cohort. Sci Total Environ 2018; 618:1046-1053

355. Giesbrecht GF, Liu J, Ejaredar M, Dewey D, Letourneau N, Campbell T, Martin JW, Team APS. Urinary bisphenol A is associated with dysregulation of HPA-axis function in pregnant women: Findings from the APrON cohort study. Environ Res 2016; 151:689- 697

356. Gao S, Casals N, Keung W, Moran TH, Lopaschuk GD. Differential effects of central ghrelin on fatty acid metabolism in hypothalamic ventral medial and arcuate nuclei. Physiol Behav 2013; 118:165-170

357. Lockie SH, Stark R, Mequinion M, Ch'ng S, Kong D, Spanswick DC, Lawrence AJ, Andrews ZB. Glucose Availability Predicts the Feeding Response to Ghrelin in Male Mice, an Effect Dependent on AMPK in AgRP Neurons. Endocrinology 2018; 159:3605- 3614

358. Engström Ruud L, Pereira MMA, de Solis AJ, Fenselau H, Brüning JC. NPY mediates the rapid feeding and glucose metabolism regulatory functions of AgRP neurons. Nat Commun 2020; 11:442

359. Stoker C, Andreoli MF, Kass L, Bosquiazzo VL, Rossetti MF, Canesini G, Luque EH, Ramos JG. Perinatal exposure to bisphenol A (BPA) impairs neuroendocrine mechanisms regulating food intake and kisspetin system in adult male rats. Evidences of metabolic disruptor hypothesis. Mol Cell Endocrinol 2020; 499:110614

360. Rosin JM, Kurrasch DM. Bisphenol A and microglia: could microglia be responsive to this environmental contaminant during neural development? Am J Physiol Endocrinol Metab 2018; 315:E279-E285

361. Cantalupo PG, Sáenz-Robles MT, Rathi AV, Beerman RW, Patterson WH, Whitehead RH, Pipas JM. Cell-type specific regulation of gene expression by simian virus 40 T antigens. Virology 2009; 386:183-191

362. Fei XC, Song C, Gao HW. Transmembrane transports of acrylamide and bisphenol A and effects on development of zebrafish (Danio rerio). J Hazard Mater 2010; 184:81-88

363. Morck TJ, Sorda G, Bechi N, Rasmussen BS, Nielsen JB, Ietta F, Rytting E, Mathiesen L, Paulesu L, Knudsen LE. Placental transport and in vitro effects of Bisphenol A. Reprod Toxicol 2010; 30:131-137

364. Zhou Q, Miao M, Ran M, Ding L, Bai L, Wu T, Yuan W, Gao E, Wang J, Li G, Li DK. Serum bisphenol-A concentration and sex hormone levels in men. Fertil Steril 2013; 100:478-482

365. Sharma N, Deshmukh R, Bedi KL. SP600125, a competitive inhibitor of JNK attenuates streptozotocin induced neurocognitive deficit and oxidative stress in rats. Pharmacol Biochem Behav 2010; 96:386-394

175

366. Bain J, Plater L, Elliott M, Shpiro N, Hastie CJ, McLauchlan H, Klevernic I, Arthur JS, Alessi DR, Cohen P. The selectivity of protein kinase inhibitors: a further update. Biochem J 2007; 408:297-315

367. Hao J, Daleo MA, Murphy CK, Yu PB, Ho JN, Hu J, Peterson RT, Hatzopoulos AK, Hong CC. Dorsomorphin, a selective small molecule inhibitor of BMP signaling, promotes cardiomyogenesis in embryonic stem cells. PLoS One 2008; 3:e2904

368. Wahid F, Shehzad A, Khan T, Kim YY. MicroRNAs: synthesis, mechanism, function, and recent clinical trials. Biochimica et biophysica acta 2010; 1803:1231-1243

369. Deiuliis JA. MicroRNAs as regulators of metabolic disease: pathophysiologic significance and emerging role as biomarkers and therapeutics. International journal of obesity (2005) 2016; 40:88-101

370. Tilghman SL, Bratton MR, Segar HC, Martin EC, Rhodes LV, Li M, McLachlan JA, Wiese TE, Nephew KP, Burow ME. Endocrine disruptor regulation of microRNA expression in breast carcinoma cells. PloS one 2012; 7:e32754

371. Avissar-Whiting M, Veiga KR, Uhl KM, Maccani MA, Gagne LA, Moen EL, Marsit CJ. Bisphenol A exposure leads to specific microRNA alterations in placental cells. Reproductive toxicology (Elmsford, NY) 2010; 29:401-406

372. Butler MC, Long CN, Kinkade JA, Green MT, Martin RE, Marshall BL, Willemse TE, Schenk AK, Mao J, Rosenfeld CS. Endocrine disruption of gene expression and microRNA profiles in hippocampus and hypothalamus of California mice: Association of gene expression changes with behavioural outcomes. J Neuroendocrinol 2020; 32:e12847

373. Mizuno TM, Mobbs CV. Hypothalamic agouti-related protein messenger ribonucleic acid is inhibited by leptin and stimulated by fasting. Endocrinology 1999; 140:814-817

374. Lee J, Morris MJ. Modulation of neuropeptide Y overflow by leptin in the rat hypothalamus, cerebral cortex and medulla. Neuroreport 1998; 9:1575-1580

375. Brakch N, Rist B, Beck-Sickinger AG, Goenaga J, Wittek R, Burger E, Brunner HR, Grouzmann E. Role of prohormone convertases in pro-neuropeptide Y processing: coexpression and in vitro kinetic investigations. Biochemistry 1997; 36:16309-16320

376. Creemers JW, Pritchard LE, Gyte A, Le Rouzic P, Meulemans S, Wardlaw SL, Zhu X, Steiner DF, Davies N, Armstrong D, Lawrence CB, Luckman SM, Schmitz CA, Davies RA, Brennand JC, White A. Agouti-related protein is posttranslationally cleaved by proprotein convertase 1 to generate agouti-related protein (AGRP)83-132: interaction between AGRP83-132 and melanocortin receptors cannot be influenced by syndecan-3. Endocrinology 2006; 147:1621-1631

377. Boucher JG, Ahmed S, Atlas E. Bisphenol S Induces Adipogenesis in Primary Human Preadipocytes From Female Donors. Endocrinology 2016; 157:1397-1407

176

378. Rezg R, Abot A, Mornagui B, Aydi S, Knauf C. Effects of Bisphenol S on hypothalamic neuropeptides regulating feeding behavior and /APJ system in mice. Ecotoxicol Environ Saf 2018; 161:459-466

379. Christiansen S, Kortenkamp A, Axelstad M, Boberg J, Scholze M, Jacobsen PR, Faust M, Lichtensteiger W, Schlumpf M, Burdorf A, Hass U. Mixtures of endocrine disrupting contaminants modelled on human high end exposures: an exploratory study in rats. Int J Androl 2012; 35:303-316

380. Hernandez Scudder ME, Kunkel MN, Gore AC. Exposure to prenatal PCBs shifts the timing of neurogenesis in the hypothalamus of developing rats. J Exp Zool A Ecol Integr Physiol 2020; 333:550-560

381. Gojska NM, Belsham DD. Glucocorticoid receptor-mediated regulation of Rfrp (GnIH) and Gpr147 (GnIH-R) synthesis in immortalized hypothalamic neurons. Mol Cell Endocrinol 2014; 384:23-31

382. Li Y, Birnbaumer L, Teng CT. Regulation of ERRalpha gene expression by estrogen receptor agonists and antagonists in SKBR3 breast cancer cells: differential molecular mechanisms mediated by g protein-coupled receptor GPR30/GPER-1. Mol Endocrinol 2010; 24:969-980

177

Copyright Acknowledgements

Permissions from journals granted for Chapter 3, Chapter 4 and Chapter 5.